With the rapid development of the nuclear energy industry, the importance of emergency management of nuclear accidents is increasingly evident. As an important component of nuclear accident emergency management, emergency evacuation is an important measure to reduce residents’ radiation risk and protect residents’ life safety. According to the requirements for the emergency plan, once the nuclear power plant releases radioactive substances to the environment, residents living in the plume emergency plan zone will evacuate from the assembly place of residents, and the terminal point is the decontamination station at the boundary of the outer of the plume emergency plan zone. In the evacuation process, residents should try their best to avoid the roads covered by radioactive plumes, so as to complete evacuation rapidly and safely. If interference events (e.g., natural disasters, traffic accidents, road construction and so on) occur at the same time, resulting in the failure of some roads in the evacuation route, evacuees will be forced to detour and the evacuation time will be increased. More seriously, they may even force residents to pass the roads covered by smoke plume, increasing the risk of residents being exposed to radiation, and causing serious consequences to evacuation. In this study, this system characteristic of road network is called road network vulnerability. It can be found that the vulnerability of the road network in the plume emergency plan zone is an important factor restricting the evacuation of residents. Therefore, this paper researches the vulnerability of the road network in plume emergency plan zone, tries to find the key road sections which restrict residents’ evacuation, and carries out targeted management and repair measures. The research results can effectively mitigate the adverse impact of interference events on residents’ emergency evacuation, so as to ensure that residents’ nuclear emergency evacuation can proceed smoothly.
Considering that road interruption has a serious impact on evacuation time and evacuation risk of residents during nuclear emergency evacuation, a method to evaluate the vulnerability of road network in plume emergency planning zone during nuclear emergency evacuation is proposed. First, the risk time index is constructed to characterize the traffic performance of the road network in the plume emergency planning zone during nuclear emergency evacuation. Next, under the mode of “government organized evacuation+self evacuation under the guidance of the government”, bus drivers are trained strictly and tend to drive along the route arranged in advance. As a limited rational person, private car owners tend to choose their own satisfactory evacuation path according to the path length, evacuation risk and traffic jam degree. Therefore, a “private car road selection rule” is proposed to plan the evacuation route of private cars. Then, based on the cell transmission theory, considering the constraints of vehicle conflict at intersections, full load of buses and road choice of self-evacuating residents, a public evacuation model is built to minimize the risk time, which is used to solve the evacuation route and risk time of residents in any wind scenarios. Then, the invalid road is selected from the evacuation routes of residents in the given wind and deleted from the road network. The resident evacuation model is resolved to get the minimum risk time of resident evacuation. The next road is selected as the invalid road and the above calculation is carried out, until all the routes for residents to evacuate in the given wind are traversed. Finally, the next wind is given and the above calculation is carried out, until all the wind directions are traversed. With wind frequency as the weight, the differences of risk time between road interruption scenario and road network integrity scenario in all wind directions are weighted summed to calculate the road network vulnerability index value and find the key road.
Taking the plume emergency planning zone road network of T nuclear power plant as an example, the vulnerability index values of road network during nuclear emergency evacuation are calculated. The five roads with higher scores are selected as key roads. Through detailed analysis, it is found that the failure of the above roads can prolong the evacuation time of residents and increase the risk of radiation exposure to residents in the evacuation process under multiple wind direction scenarios. The results show that the proposed method can effectively assess the vulnerability of road network in plume emergency planning zone during nuclear emergency evacuation, identify the weak links of the network, and provide scientific basis for the maintenance and reconstruction of road network in plume emergency planning zone and nuclear emergency management.

In recent years, the outbreak of sudden disasters such as corona virus pneumonia, floods, earthquakes, and fires has occurred frequently, causing huge casualties and economic losses to our country. In this context, to save human life as much as possible and minimize the loss, it is particularly necessary to supply emergency materials quickly and in time. Therefore, the location-routing optimization problem of emergency logistics distribution center is worthy of in-depth study.
By sorting out the existing research literature, domestic and foreign scholars have carried out a lot of research on logistics distribution center location, distribution route optimization and joint location-routing optimization, but there are still two research gaps: Firstly, most of the existing literature is based on static deterministic network to carry out problem analysis and modeling. The speed of delivery vehicles is constant, but the traffic flow in the real urban road network has significant differences in time and space distribution, and traffic conditions have time-varying characteristics. The influence of the traffic network with real-time speed changes on the emergency facility location-routing optimization problem is ignored. Secondly, in the algorithms for solving the emergency facility location-routing optimization problem, such as the branch and bound method, Dijkstra algorithm, dynamic programming method, etc., the time and space complexity of the algorithm are large, and the calculation rate is low, such as ant colony algorithm, genetic algorithm, tabu search algorithm, simulated annealing algorithm, variable neighborhood search algorithm, quantum competitive decision-making, simulated plant growth and other intelligent optimization algorithms. The applied research on the location-routing optimization problem under the condition of time-varying road network is still insufficient, and the solution accuracy needs to be improved.
To sum up, the time-varying traffic network is rarely considered in the previous research on the location-routing optimization problem of emergency facilities, and the algorithm for solving it has shortcomings such as low calculation rate and poor solution accuracy. In emergency management, time is life. In the road traffic network, the time-varying characteristics of the path travel time caused by a change in the traffic flow throughout the day, and the different travel speeds in the road traffic network will affect the timeliness of emergency rescue. The location selection will also affect the travel time. Therefore, it is necessary to develop an accurate search algorithm that can effectively overcome the time-varying characteristics and uncertainties of dynamic traffic networks and provide technical support for the rapid solution and practical application of the model.
In view of this, this paper uses a multi-stage algorithm based on a time-varying traffic network to solve the location-routing optimization problem of emergency logistics centers, namely the ripple spreading algorithm, which evolves synchronously with the change in the time-coordinated environment of each unit. In each calculation process, the ripple relay race splits many ripples with the change in the node state, and the ripple spreading speed is consistent with the speed of the current road segment in unit time, reflecting the co-evolution of theoptimization steps and the road network traffic environment, generating a set of independent ripple relay race to explore the global optimal path. Taking the vehicle speed of different road conditions in different time periods into consideration, the co-evolution of the emergency vehicle distribution process and the time-varying speed is realized, and the site selection-path optimization that minimizes the total distribution time from the alternative emergency logistics center to all target nodes is constructed.
Taking an emergency material reserve center in Beijing as an example, the application of the above model is demonstrated, using the ripple spreading algorithm, the genetic algorithm, the dynamic Dijkstra algorithm and the branch-and-bound method to solve the optimal location-routing optimization scheme respectively. The calculation speed is much faster than the other three algorithms. In terms of the optimality of delivery time, the optimal solution can be obtained like the dynamic Dijkstra algorithm and the branch and bound method. Although the genetic algorithm is faster, the solution accuracy is very low. The speed change in the time-varying road network will lead to the change in the optimal emergency logistics center solution. The distribution path of the ripple spreading algorithm, and the real-time change of the traffic road network speed are more synergistically coupled and can be used in large-scale site selection. The optimal or better solution is achieved in a short time, and it exhibits excellent optimality and robustness in solving complex time-varying network location problems.

Various natural disasters, accidents, and events that endanger public health or security have occurred throughout the world over the past few decades. An appropriate emergency shelter location can decrease the losses the disasters bring. Due to the uniqueness of emergencies, uncertain information should be considered in location decision making. The uncertainty of vertex weight has been studied in many related researches. Taking the uncertainty of edge capacities into consideration, this paper studies the emergency shelter location problem in a dynamic path network. With a given dynamic path network, each vertex is assigned a positive weight which indicates that the number of people should be evacuated, and each edge has a positive length and an uncertain capacity. The edge capacity is the maximum number of people that can enter the edge per unit time. For each edge, although the capacity is not known exactly, the interval to which it belongs is given. A vertex probably gets congested if too many people try to get in the corresponding edge at the same time. The time spent due to congestion should be noted during the evacuation. The problem requires a location so that the maximum regret of the maximum completion time is minimized.
The first part of this paper gives basic definitions of the problem and some useful properties are also presented. The capacity of each shelter (sink) is assumed to be infinite. The evacuation is completed as soon as the weight reaches a shelter, so if a shelter is located exactly on a vertex, then all the weight of this vertex can complete evacuation with no time. Based on the uncertainty of edge capacity, congestion situation during the evacuation progress is analyzed in detail, and then critical edge capacity leading to changes in congestion situation is successively calculated, so we can get the maximum completion time. With a given scenario, that is to say, the capacity of each edge is given, the maximum completion time of the left or the right part from a sink point is proved to be unimodal with the sink location moving on a path network. Then the effect of critical edge capacity on the maximum completion time is analyzed.
In the second part of this paper, a polynomial algorithm to solve the sink location problem on a dynamic path network with uncertain edge capacities is proposed. First, based on the congestion changing during the evacuation, the marginal value of edge capacity that causes a change in the last congestion point is analyzed. The special structural characteristics of the maximum regret scenarios are analyzed to ascertain the worst case scenario for all possible maximum regret values. And the number of all possible worst case scenarios is restricted to polynomial. Then, all the locations located on vertices and edges are treated separately. To find the min-max regret sink location, linear programming model is applied in this part.An O(n³)-time algorithm is developed to solve the 1-sink location problem. Then the uncertainty of vertex weight is taken into consideration at the same time further, and an O(n⁵)-time algorithm is proposed.
A practical example is presented in the third part of this paper. The practical example is proposed based on Xianyang Lake Scenic Spot which is abstracted to a path network with 12 vertices. Compared with present emergency shelter location, the new location has a good performance when the worst case scenario happens. This result can provide a theory gist for emergency shelter location and network building.
In summary, this paper studies the min-max regret 1-sink location problem on a dynamic path network with interval edge capacity. To obtain the maximum completion time, the dynamic congestion progress of all the weight during the evacuation is analyzed in detail. It is proved that the maximum completion time is unimodal. Although the number of edge capacity scenarios is infinite in theory, this paper limits the possible worst case scenarios to polynomial, which is based on the special structural characteristics of the maximum regret scenarios. For each edge, the min-max regret sink point is obtained by applying linear programming model. Considering all the locations on vertices and edges separately, the algorithm proposed in this paper gets a time complexity of O(n³). An O(n⁵) time algorithm is also proposed to solve this problem when the vertex weight is also given as interval data. In our future research, the multiple sink location problem will be considered. In addition, this problem will be also studied with more general dynamic networks, such as trees and general connected networks.

In recent years, major emergencies have occurred frequently all over the world. The scheduling of emergency medical supplies is the key link of emergency medical supplies support. There are some issues of the single-cycle multi-transportation mode emergency medical supplies scheduling model, such that the transportation network is not appropriate, constraints of transportation modes are not considered comprehensively, or dynamic changes of supply and demand are not considered, resulting in local redundancy and unfairness of scheduling.
Firstly, the multi-period dynamic scheduling model of emergency medical supplies under emergencies is established. According to the response characteristics of emergency medical supplies after the outbreak of emergencies, this thesis adopts the three-level transportation network, from supply point to distribution center and to demand point. It uses a variety of transport tools to transport jointly, considering capacity constraints for trucks and departure schedule constraints for trains and aircraft. It establishes a dynamic demand urgency indicator for demand points and calculate a dynamic demand urgency of demand points to correct their satisfaction rates. By dividing the period and aiming at the shortest scheduling time and the maximum sum of the satisfaction rate of the affected people, a model of emergency medical supplies is constructed with multi-supply point, multi-distribution center, multi-demand point and multi-period dynamic scheduling.
Secondly, an algorithm based on non-dominated sorting genetic algorithm-II (NSGA-II) is designed for solving the dynamic scheduling model of emergency medical supplies. For individual coding, the combination of string coding and natural number coding can intuitively show the entire scheduling scheme. For non-dominated sorting, it is difficult to consider all constraints in individual coding due to a large number of model constraints. The traditional non-dominated sorting method is improved, and the individuals who meet the demand of the constraints and those who do not are hierarchically divided according to the magnitude of individuals exceeding the constraints.
Finally, a case of multi-period dynamic scheduling of emergency medical supplies is studied. The designed case is solved by the NSGA-II algorithm. The results show that at the end of scheduling response of emergency medical supplies, the higher the demand urgency of demand point is, the more its satisfaction rate exceeds the minimum satisfaction one. With an increase in the satisfaction rate of demand point in the whole scheduling process, the final demand is satisfied, which verifies the validity and rationality of the model. At the same time, the genetic algorithm (GA) is used to compared with the NSGA-II algorithm. The results show that the scheduling scheme set solved by the NSGA-II algorithm dominates the scheduling scheme by the GA algorithm, reflecting the effectiveness and advantage of the NSGA-II algorithm.
The conclusions show that the model built in this paper can consider the constraints of different transportation modes, and materialize the idea that the higher the urgency of the demand point, the greater the value of its satisfaction rate exceeding its minimum satisfaction rate, which ensures the fairness of the distribution of emergency medical materials.

Programs are the basic form of achieving multi-level strategic goals and also the basic unit for implementing major projects. In the process of program implementation, program delay management plays an important role in the overall implementation of the program. Compared with a single project, contracted project delays under multiple stakeholders bring two more significant adverse effects to the program: firstly, there are more factors that affect the on-time completion rate of contracted projects in the program, and the chain reaction is more significant, which brings adverse effects to other contracted projects, making the control of other contracted projects more complex and difficult; secondly, according to the current rules of delay compensation and claim contract terms, most of the risks brought by contracted project delays to the program are borne by the employer. This means that the adverse effects of contracted project delays on other contracted projects and peak shaving are not only nonlinear, but also have a chain and amplification effect in a multi stakeholder environment. Therefore, starting from the perspective of stakeholders, how to balance the needs of the employers while reducing the impact of contracted project delays on the entire program, thereby reducing the risks and responsibilities of the employers, is a problem that needs to be addressed.
This article, in accordance with the current rules on compensation and claims for delay damages (such as FIDIC), reveals the mechanism of cost increase and changes caused by contracted project delays to employers, and introduces the critical chain method to construct a nonlinear optimization model for program delay costs based on the critical chain. This enriches the theory of program cost optimization, and by effectively monitoring the buffer of the critical chain program, the balance of NCRPE, start rate of contracted projects, and on-time completion rate of milestone schedules and program schedules can be improved, to reduce the negative impact of NCRPE imbalance and contracted project delays on the program from the source, so that employers can assess their own risks. At the same time, employers can achieve a balance among program duration, program delay cost, NCRPE demand intensity balance coefficient, and maximum NCRPE demand intensity through key chain buffer settings based on their employer management preferences.
The first part elaborates on the non-linear relationship between the cost increase caused by contracted project delay and time in three aspects: firstly, there is a non-linear relationship between the losses caused by contracted project delay to the contracted project itself and other contracted projects and time; secondly, there is also a non-linear relationship between the additional costs and time caused by the delay of other contracted projects in the program and the requirement to compress their construction period; the third is the nonlinear relationship between the NCRPE supply imbalance coefficient and cost, and the nonlinear optimization principle of program delay cost based on critical chain is proposed.
The second part starts from the two stages before and during the implementation of the program, and constructs a nonlinear optimization model for program delay cost based on the critical chain. Considering the different joint delay effects caused by contracted project delay caused by distance on subsequent contracted projects, PERT is used to calculate the delay probability of contracted projects, and the joint delay effects caused by contracted project delay are derived, integrating this nonlinear impact into critical chain buffer monitoring, reducing the adverse effects of contracted project delays by identifying critical chains, allocating buffers, conducting buffer monitoring, and taking corresponding corrective measures.
The third part starts from two aspects: the general preference and extreme preference of the employer for the construction period. Through example analysis, the feasibility and effectiveness of the constructed model are verified. Before the implementation of the program, the peak shaving method is used for NCRPE equilibrium optimization. During the implementation of the program, the constructed model is applied to optimize the construction period, delay costs, etc. Finally, management insights for practical reference are proposed for both the employer and contractor.
Through simulation and calculation, the results show that the program’s duration, increased delay cost, maximum imbalance coefficient, and maximum demand intensity are related to the employer’s preference for duration. Compared to the initial state of the program, to some extent, this model can reduce the program’s duration, increased delay cost due to contracted project delays, maximum imbalance coefficient, and maximum demand intensity. However, compared to the optimization state before the implementation of the program, due to considerations of schedule and cost, a certain maximum demand intensity may be sacrificed.
There are two directions for further research: firstly, from the perspective of contractors, with the goal of maximizing their own interests, we can adjust the contracted project plan under constraints. The second is to conduct research on the nonlinear impact of uncertain events under various constraints on the program and its components.

With the rise of artificial intelligence, human-robot collaboration has significantly improved the efficiency and quality of tasks. As the main participants of human-robot collaboration, humans play a key role in collaboration. The current researches on the influencing factors of human-robot collaboration performance are mainly carried out from the two aspects: (1)The influence of robot characteristics on the collaboration performance, including the characteristics, adaptability and degree of automation of robots. (2)The influence of human-robot relationship on collaboration performance, including human-robot trust and shared mind between human-robot, etc. The above researches have considered the influence of robot behavior characteristics on human-robot collaboration performance. However, few studies have explored the influence of human subject traits on human-robot collaboration performance with human as the center. In addition, the researches on human-robot relationship mainly focus on the trust relationship between human and robot, but those on human-robot collaboration and interaction under different robot characteristics are particularly lacking. Therefore, the scientific questions studied in this paper are: What is the influence mechanism of human rationality degree on human-robot collaboration performance? How does this influence mechanism differ in different robot adaptations?
Due to the dynamic and adaptive nature of human-robot collaboration, the existing methods based on physical experiments and questionnaires take cost and time, and are difficult to understand and reveal the evolutionary characteristics of collaboration from a dynamic perspective. Therefore, based on the bounded rationality theory, this paper integrates the virtual game model and reinforcement learning algorithm to establish an agent-based human-robot collaborative computational model considering human rationality, and then, to observe the influence of human rationality degree on objective and subjective performance of collaboration. Objective performance refers to the completion of objective cooperative tasks between human and robot, while subjective performance refers to the subjective evaluation of collaboration by human collaborators. In this paper, Python language is used to program the model and conduct simulation experiments. To avoid the influence of randomness, experiments are used to run 100 times in each experiment with 200 cycles in each time, and the average value of 100 experiments is used for indicator measurement.
Through a simulation experiment of a move collaborative task, the influence of human rationality on collaborative performance and the moderator role of robot adaptability are analyzed. The experimental results show that increasing the degree of human rationality can improve the performance of human-robot collaboration, but this effect marginally diminishes for objective performance. Besides, the higher the adaptability of robots, the more obvious the impact of human rationality on collaboration performance. The research extends the bounded rationality theory to the field of human-robot collaboration performance improvement. And the research conclusions have practical enlightenment for the management of human-robot teams. Improving the cognitive ability of human collaborators, adopting robots with high adaptability and recruiting human groups with a high degree of rationality will lead to dramatic improvements in the overall performance of human-robot collaboration.
The simulation scenario in this paper is based on mobile cooperative tasks. With the diversification of human-robot collaboration scenarios, we will try to further explore other types of human-robot cooperative tasks in the future. In addition, agent-based computational experiment is a simplification of the real world on the basis of ensuring the rationality of the model. Moreover, it cannot truly represent the complex decision-making process of human beings, such as the influence of psychology, physiology, environment, etc., on decision-making. Behavioral experiments involving human beings can be considered to further demonstrate the conclusions obtained in this paper.

Due to the popularity and rapid development of cloud computing, many Internet companies have continued to build ultra-large-scale data centers for the past years. When receiving a new virtual machine request, the cloud computing center searches for a suitable physical machine and creates a virtual machine with the corresponding specifications. When the customer finishes using the virtual machine, the cloud computing center will delete the virtual machine and free the corresponding space. As time goes by, the remaining space of each physical machine in the computer room cluster will vary greatly, resulting in the waste of physical machine space and an increase in fragmentation rate. Therefore, the cloud computing center needs to periodically readjust the virtual machine deployment scheme (rescheduling) to improve the utilization of physical machine resources. Since the hot migration operation consumes more communication resources and operation time, it is necessary to minimize the impact of deployment differences on physical machines during rescheduling to achieve optimal adjustment plan.
Many studies have focused on the scheduling of virtual machines in static scenarios, but the frequent creation and deletion of virtual machines in the cloud computing center leads to a waste of physical machine space, which relies on virtual machine rescheduling to improve the operational efficiency of the cloud computing center. After carefully searching, we find that no scholars have studied the virtual machine rescheduling problem. Therefore, considering the important impact of rescheduling in virtual machine scenarios, this paper studies the virtual machine rescheduling problem in cloud computing centers. Given the existing deployment plan of the virtual machine and users’ virtual machine requirements, it readjusts the virtual machine to meet the actual needs of the user. The key lies in minimizing the total cost, including operating costs and deployment variance costs, and improving the efficiency of cloud computing center resource utilization by optimizing the virtual machine rescheduling scheme. The research results of this paper are helpful for improving the flexible deployment capability of cloud computing centers and better coping with the market competitive environment.
We first present the mathematical definitions of placement deviation and then formulate the virtual machine rescheduling problem as a mixed integer mathematical model. The objective is to minimize the total costs including wasted resources, deployment mode preparation costs, and variance costs. The virtual machine rescheduling scheduling problem introduces differential variables compared with the static deployment problem and is more difficult to be solved. According to the characteristics of the problem, we propose an accurate algorithm based on branch and pricing. Firstly, the Dantzig-Wolfe decomposition method is used to rewrite the model of the rescheduling problem, then the column generation is used to obtain the optimal solution of the relaxation problem, and finally the branch and price approach is used to obtain the optimal solution of the problem. The branch and price algorithm can accurately solve the virtual machine rescheduling problem, but it will take a long time to solve large-scale or enterprise instances, which is not suitable for industrial applications. Therefore, we also propose a heuristic search strategy based on column generation to achieve a high-quality rescheduling scheme in a short time.
In this paper, numerical experiments are carried out by randomly generating examples to evaluate the performance of the proposed Branch & Price algorithm and CGH algorithm. For small-scale examples, the Branch & Price algorithm can obtain the optimal solution of all examples, and the average gap of CGH algorithm is 1.02%; For medium and large-scale studies, for some examples, CPLEX can no longer obtain the optimal solution within 1 hour, but the Branch & Price algorithm still finds the optimal solution of all examples, and the gap of CGH algorithm is only 1.98%. Finally, a sensitivity analysis is carried out for different cost parameters to analyze the influence of parameter changes. The results show that the decrease in the pattern setup cost and deviation cost respectively increase the corresponding number of patterns and deviation patterns, but decrease the total cost. The cloud computing center can smooth pattern deviation by acknowledging customers’ demand in advance and improving flexible placement capability, and further promote operation competitiveness.

With the development of global economic integration, manufacturing enterprises have to improve customer service levels from the perspective of the overall supply chain. Production and distribution are two important decisions for make-to-order (MTO) manufacturers. These two decisions are generally made in a sequential manner, in which production schedules are firstly generated by the production department and the transportation routes are subsequently provided by the logistics department. The lack of joint optimization of production scheduling and distribution may increase the operation cost but decrease the customer satisfaction, especially in time-intensive industries, such as newspapers, perishable products, and fashion garments, etc. In addition, to further enhance cost competitiveness and market responsiveness, distributed manufacturing has become a general trend for large manufacturing enterprises because of its higher production profits, lower management risks, and faster manufacturing cycles. Therefore, the coordination of production and distribution among distributed factories has recently attracted an increasing attention in both manufacturing industries and academic community.
This paper focuses on a coordinated distributed hybrid flow shop scheduling and vehicle routing problem (CDHFSVRP) for minimizing the sum of transportation cost and delay cost. A search of available literature shows that no research efforts have been devoted to addressing the CDHFSVRP. The CDHFSVRP consists of two combinatorial optimization problems, namely the distributed hybrid flow shop scheduling problem in the production stage and the vehicle routing problem in the distribution stage. To address the CDHFSVRP, four types of decisions have to be made, namely factory allocation, determination of job processing sequence on machines, vehicle allocation, and distribution route selection. Considering the NP-hardness of the CDHFSVRP, a hybrid algorithm (EDA-ALNS-DQ) that integrates a distribution estimation algorithm (EDA) and an adaptive large neighborhood search (ALNS) is presented to generate coordinated schedules. In this algorithm, EDA firstly employs the probabilistic models to generate populations, and then ALNS with six types of destroy operators and three types of repair operators further improves some good solutions. Although the traditional ALNS applies the well-known roulette wheel selection method to determine the operators, it will be difficult to identify an appropriate operator when most of the considered operators provide similar performance. To enhance the local search performance of ALNS, Q-learning is adopted to exploit the knowledge of solution space and accordingly select the operators based on the reward matrix.
To evaluate the performance of the proposed hybrid algorithm, two sets of test problems are randomly generated based on the previous study of ULLRICH’s. All the compared algorithms are coded with C++ and run on a PC with Intel CoreTM i5-1035G1 CPU 1.00GHz processor. EDA-ALNS-DQ is firstly compared with CPLEX on a set of 18 small-sized problems. Both CPLEX and EDA-ALNS-DQ reach the optimal results of 13 problems, and EDA-ALNS-DQ spends much shorter CPU time. For the other 5 problems, CPLEX fails to obtain the optimal solutions in a predefined computation time, whereas EDA-ALNS-DQ is capable of offering better solutions, each of which is equal or less than the upper bound of CPLEX. Furthermore, EDA-ALNS-DQ is evaluated on a set of 27 large-sized problems, and it statistically performs better than some competitive scheduling algorithms, namely traditional EDA, ALNS, and EDA-ALNS. EDA-ALNS is the same as the proposed EDA-ALNS-DQ, except for determining the destroy and repair operators using the traditional well-known roulette wheel selection method. As a novel meta-heuristic, EDA-ALNS-DQ is capable of making a good balance of search exploration and exploitation, and therefore provides a promising approach to integrating distributed hybrid flow shop scheduling and vehicle routing.
The proposed EDA-ALNS-DQ is expected to help operation managers of manufacturers to make better decisions on the coordination of production and distribution among distributed factories. Future research may focus on integrated production and distribution in more realistic industrial environments, such as the consideration of more complicated production environments, uncertain transportation time, and multiple types of transportation modes, etc. Another potential research area is to develop more effective heuristics or meta-heuristics to deal with the CDHFSVRP.

Sparrow search algorithm (SSA) is easy to fall into the local optimal point in the search process, resulting in a single population diversity in the late iteration stage, so multi-strategy sparrow search algorithm (MSSSA) is proposed in this paper. In practice, random number sequences are widely used to generate the initial population in the search space, but the distribution uniformity of the initial population by this method is not good, and the exploratory nature of the initial population to the solution space cannot be guaranteed.
Firstly, the low-discrepancy sequence and chaotic mapping are used to generate a uniform initial population to ensure the uniformity of the initial population, so as to accelerate the convergence rate of the algorithm and improve the ability of the algorithm to jump out of the local optimal value. The producer of the SSA, as the main means of exploring the search space of the algorithm, does not balance local search and global search well. In order to solve this problem, the Euclidean distance and nonlinear inertia weight information between the current sparrow position and the upper and lower limits of the search are fully used to balance the accuracy and breadth of the producer’s search. In order to improve the ability to jump out of the local optimum, the random unit vector with uniform distribution of the d-dimensional hypersphere is introduced, which fully improves the random walking nature of the population.The scroungers strategy of the SSA makes most of the followers only follow the current optimal sparrow, sacrificing a certain global search ability, and is easy to fall into the local optimal value.
For preventing the loss of the current optimal search information, the elite strategy is introduced to preserve the optimal individual and improve the convergence speed and accuracy. When the former optimal individual is too close to the coordinate origin, the Gaussian variation range is limited, and it plays a very limited role in the process of jumping out of the local optimum. Although the reverse search strategy of one-dimensional oppositional learning reduces the interference between dimensions and expands the search area of the algorithm, it is difficult to improve the optimization accuracy of the algorithm in the late stage of population iteration. Therefore, the Gaussian mutation method and the one-dimensional oppositional learning are comprehensively improved, and the improved Gaussian mutation and improved one-dimensional oppositional learning are used to perturbate and accurately search at the position of the optimal solution, so as to improve the performance of the algorithm.
Under the same experimental conditions, the proposed algorithm (MSSSA) is compared with 9 algorithms on 24 benchmark functions, 50-dimensional and 100-dimensional CEC2017 test functions, and the excellent performance of the proposed algorithm in optimization ability is proved by the Wilcoxon signed ranks test. In order to verify the effectiveness of the comprehensive strategy, model ablation experiments are carried out on the algorithm, and the comprehensive effectiveness of the proposed improved strategy is proved. The comparison with the practical engineering application of 9 algorithms in the classical pressure vessel design problem also shows that the MSSSA has better optimization ability, effectively saves production costs, and has certain practical application value.

The defense intellectual property cooperation under the background of civil-military integration is a joint investment of superior resources from both military and civilian enterprises. The cooperation aims to develop the defense intellectual property that meets their needs, which in turn promotes the two-way development of national defense construction and national economy. How to encourage and deepen the defense intellectual property cooperation under the background of civil-military integration, and solve the problems such as unclear ownership of property rights, barriers to information communication, and disputes in interest distribution of cooperation are crucial to promoting the in-depth development of civil-military integration.
Current studies on defense intellectual property cooperation are limited to qualitative analysis, lacking the construction of the evolutionary game model on defense intellectual property cooperation under the background of civil-military integration. Therefore, we explore the cooperation mechanism of defense intellectual property in the context of civil-military integration by constructing a tripartite evolutionary game model with military firms, civilian firms, and the government as game subjects. Several typical characteristics of civil-military integration, such as the national defense benefits, risk aversion, and information communication costs are included in the tripartite evolutionary game model.
The experimental analysis process is as follows: First, we clarify the ownership of defense intellectual property and put forward hypotheses on the cooperation. Second, we construct the tripartite benefit matrix of the evolutionary game model to establish the replicator dynamics equations. Third, according to the replicator dynamics equations, we obtain the Jacobian matrix and obtain the eigenvalues of Jacobian matrix. Finally, we analyze the evolutionary stability strategies and investigate the critical factors affecting the stability of cooperation through numerical simulation. We discuss the evolutionary state of the system under different government subsidy coefficients, income distribution coefficients, default penalties, and information communication costs.
The results show that the stability of defense intellectual property cooperation is positively correlated with the penalty for breach of contract and negatively correlated with the information communication cost. And there is an inverted U-shaped relationship between the stability of cooperation and the income distribution coefficient. The subsidy and punishment of the government are crucial to the stability of cooperation. Military firms are more sensitive to government policies and the penalty for breach of contract, while civilian firms are more sensitive to the information communication cost. We will also determine the effective interval of the income distribution coefficient and the government subsidy coefficient when the state of the optimal game strategy is stable.
Based on the evolutionary game model, this study broadens the research methods of defense intellectual property cooperation under the participation of the government. At the same time, according to the characteristics of defense intellectual property cooperation under the background of civil-military integration, we include the national defense benefits, risk aversion, information communication costs, as well as the government’s social benefits and loss of credibility into the evolutionary game model, which is an extension of the existing evolutionary game model on the intellectual property cooperation in the field of civil-military integration. Moreover, we consider the government’s penalties for breach of contract, which is ignored in current evolutionary game models on civil-military integration.
From the aspects of government subsidies, income distribution, default penalties, and information communication, corresponding countermeasure suggestions are proposed for improving the stability of cooperation. The government should reasonably adjust the fund subsidy coefficient and tax support coefficient. To improve the willingness to cooperate, the government can formulate corresponding incentive policies, provide certain tax incentives and strengthen intellectual property protection. Civilian and military firms should determine a reasonable income distribution coefficient, objectively and accurately measure the contribution of the enterprise from various aspects, and establish an effective interest balance mechanism. Both enterprises and governments should provide effective punishment mechanisms for breach of contract to maintain the loyalty of cooperative subjects and enhance mutual trust through better communication. Moreover, we should promote the construction of information-sharing platforms between military and civilian firms, in order to reduce the cost of information communication and promote technology integration and innovation. Military enterprises can appropriately open the information at a low secret level to civil enterprises, and timely decrypt the technical information that is not currently classified, so as to promote the conversion of military technology to civilian use.
However, there are some limitations in our research. First, we do not consider the attitude of military and civilian firms toward different risk types. The cost of risk can be introduced to reflect the attitude of risk, and the degree of risk avoidance to different risk types can be described according to the size of different risk costs. Second, we do not consider universities and scientific research institutions, which can also participate in the research and development of defense intellectual property. Therefore, universities or scientific research institutions can be included as the subjects of the gaming model in subsequent studies for further research.

With the rapid development of the world economy, the production mode aiming at maximizing economic benefits has gradually deteriorated the ecological environment and seriously threatened the sustainable development of human society. In this context, green technology innovation has received extensive attention from scholars and industries at home and abroad as a fundamental way to solve the dilemma between the economic development and environmental pollution. As a systematic project, green technology innovation is usually completed by multiple stakeholders. There are great differences in the interest needs of various stakeholders, and they restrict and influence each other and work together on the effect of green technology innovation. Therefore, it is particularly important to clarify the game relationship between stakeholders in green technology innovation, and the mechanism and process of game evolution. The research enriches the research results of green technology innovation and expands the depth and breadth of the theory of collaborative innovation. At the same time, it provides ideas for relevant government departments to formulate relevant policies to improve the collaborative innovation environment of green technology innovation and provides theoretical support for enterprises and research institutions to formulate cooperative innovation strategies and improve the performance of industry-university-research cooperation in green technology innovation.
Based on the stakeholder theory and evolutionary game theory, we construct a four party evolutionary game model for green technology innovation involving government, enterprises, research institutions, and the public. We analyze the influencing factors of the strategy selection of each game subject in the game system and the behavior mechanism of each subject. With replicator dynamic equations to obtain the eigenvalues of the Jacobian matrix for pure strategy equilibrium points, the stability of the equilibrium points in the system is analyzed based on the Lyapunov first rule. Through the simulation analysis of Matlab software, on the one hand, the above analysis results are verified, and the evolution path of the four-party game system of green technology innovation is more intuitively seen. On the other hand, the influence of key parameters such as government punishment, cost-sharing coefficient, and cooperative innovation income on the evolution process and results of the system is quantitatively investigated.
The results shows that: there are 10 conditional stable points in the 16 pure strategy equilibrium points of the evolutionary game system, among which (0,1,1,1) are the most ideal stable points of the system. According to the simulation results, it can be seen that a reasonable cost allocation mechanism is the basis for the achievement and stability of green technology collaborative innovation; the government’s appropriate increase in punishment is conducive to the achievement of collaborative innovation; the more significant the cooperative innovation income, the more conducive to the stability of the cooperation between enterprises and research institutions and the promotion of public participation in green technology innovation. Reducing the cost of public participation can promote the achievement of collaborative innovation among enterprises, research institutions, and the public. Based on the simulation results, some countermeasures and suggestions are put forward, such as establishing a reasonable cooperative cost-sharing mechanism and a reasonable reward and punishment mechanism, and establishing and improving public participation in green technology innovation channels, to provide a theoretical basis and policy reference for the development of green technology innovation in China.
Because it is difficult to obtain the relevant data on green technology innovation, the parameters in the simulation are not the actual parameters, which can only reflect the behavior trend of each subject in the evolutionary game system. Therefore, in the follow-up research, we will actively conduct an in-depth research on enterprises, universities, scientific research institutes, and other relevant units, and use real data for further research, so that the theory can be effectively combined with the application.

The basin resources are shared by all places along the line for their special status as public goods. For transboundary rivers, they are shared and managed by all participants in the basin. However, there are differences in geographical location, economic development and dependence on water resources among participants. Therefore, different participants have different degrees of attention to river pollution control and expected investment, which are contradictory. The disunity of upstream and downstream countries’ pollution control intention is the key factor hindering the pollution control of transboundary rivers. The pollution control of transboundary rivers has the following two characteristics: On the one hand, the fluidity and spatial timing of rivers lead to the negative externality of the pollution control of transboundary rivers, that is, the pollution behavior of one country or region will have adverse effects on the downstream countries. In contrast, downstream participants are at a disadvantage, so this conflict of interests often leads to tension and conflict between upstream and downstream participants. On the other hand, the non-competitive nature of pollution treatment of transboundary rivers means that less treatment of river pollution by countries along the basin may directly affect their own water usage. Upstream participants are not significantly negatively affected by the quality of water usage in their country by non-participation or low participation in pollution control. Cross-border rivers are shared by all parties, and the quality of their water is of vital interest to all. Therefore, whether good international cooperation can be reached in the process of river pollution control and cross-border river joint management is the key to improving the water quality of the basin.
Considering the spatial differences among participants of transboundary rivers, this paper uses the first-order expression equation of the evolution of concentration value of organic COD natural digestion in the basin to describe the relationship between river pollution control and river spatial timing. Four cost allocation methods are selected according to the development law of alliance formation for comparison, that is, initial alliance formation, dynamic trial-and-error adjustment, bringing in more partners and maintaining stability. The minimum core method that emphasizes minimum cooperation cost, SCRB method that focuses on dynamic process of decreasing cost, Shapley value method that emphasizes fairness, and Owen value method that focuses on stability are compared, aim to comprehensively consider multiple factors such as fairness, attractiveness, stability, strategic adaptability and sustainability. To develop more effective strategies for pollution control of transnational rivers, this comparative analysis helps decision-makers better understand the advantages and disadvantages of various approaches to better meet the needs and interests of all parties to improve water quality and environmental sustainability. Finally, a case study of Lancang-Mekong River is carried out to explore a win-win cooperation system for trans-boundary river pollution control from the perspective of cost allocation.
For watershed pollution control, the total cost of cooperation is significantly lower than that of non-cooperation. The cost sharing among the participants is basically based on the difference in the individual cost of pollution control among the participants, and the results of the middle tier countries (Thailand and Cambodia) are relatively different with several methods of sharing. Last-resort countries share the least in a cost-sharing approach that takes into account marginal contributions. When conducting joint pollution control in the Lancang-Mekong River basin, full consideration should be given to the difference in marginal cost of cooperation among participants and the amount of pollution emitted by participants in the pollution process, which is the focus of basin cooperation in pollution control and runs through the whole process of cost sharing in the cooperative alliance. Fairness, stability and satisfaction of cost sharing are the key to international basin pollution control.

With the continuous improvement of the level of global industrialization, the large amount of greenhouse gas emissions has led to an increasingly prominent climate problem. Facing a severe environmental pressure, on the one hand, the national carbon emission trading market has been officially launched, which combines low-carbon compensation and high-emission constraints. On the other hand, the improvement of environmental protection awareness makes consumers more inclined to low-carbon consumption, which provides impetus for the generation of low-carbon operation models of supply chain production and sales enterprises. However, the benefits and costs of carbon emission reduction in the supply chain are difficult to match, and the existence of positive external effects further hinders the smooth implementation of low-carbon supply chains. Therefore, how to use the government’s environmental regulation measures to encourage sellers and manufacturers to interact and restrict each other through the carbon trading market and product market is the key to exploring the stable development of low-carbon supply chains in the context of consumers’ low-carbon preferences.
Based on the fact that the conditions of complete rationality and complete information of traditional game participants are difficult to meet, the evolutionary game theory, which is assumed to be more realistic, is used by more and more scholars to analyze the carbon emission reduction behavior of supply chain member companies under the carbon trading mechanism. Based on the bounded rationality assumption of the game subject, this paper accurately describes the behavior of the participating players by setting different parameters, constructs a tripartite evolutionary game model composed of the government, retailers and manufacturers, and obtains a replicative dynamic system. This paper first analyzes the strategic stability of a single player with the stability principle of differential equations to study the relationship between the strategic choice of a single player and the decisions and parameter changes of other players, and then uses evolutionary game theory to conduct a holistic analysis of the three-party evolutionary game system. Finally, the interaction mechanism of game players and the influencing factors of low-carbon production and marketing mode are further explored with the help of MATLAB numerical simulation in order to find the optimal game equilibrium strategy and evolution path.
This paper draws the following conclusions: (1)During the operation of the national carbon trading market, achieving the stable development of low-carbon supply chain is the result of multi-stakeholder game interaction. (2)Carbon trading policies can effectively improve the enthusiasm of retailers and manufacturers to form a combination of low-carbon production and marketing strategies. When the policy incentive effect is limited, the government will need to improve the intensity of low-carbon subsidies and penalties to ensure certain environmental regulation effects. (3)When the probability of carbon theft of traditional manufacturers increases, the incentive effect of carbon trading policy on low-carbon transformation will be weakened, but the policy effect of punishment regulation will be strengthened. (4)Consumers’ low-carbon preference is mainly reflected in the impact of carbon emission reduction level and promotion level of low-carbon products on demand. The increased market demand will promote supply chain members to make low-carbon decisions. Based on the above conclusions, the following suggestions are obtained: (1)The government should maintain and gradually improve the national carbon trading market through various measures, and give full play to the important role of carbon trading policies in guiding the stable development of low-carbon supply chains. (2)When carbon trading and government environmental regulation are in parallel, regulatory decision-making should do a good job of supporting carbon trading policies. For supply chain companies that actively participate in low-carbon production and sales, the government should reduce their low-carbon costs through various forms of support. For companies that insist on traditional production or even carbon stealth emissions, the government should deter their luck by severely punishing them. (3)The government and enterprises should strive to guide consumers’ low-carbon awareness, create an environment conducive to low-carbon consumption, and finally realize the ideal scenario of the carbon trading market and the product market jointly promoting the low-carbon production and marketing of the supply chain from the perspective of government environmental regulation.

The intertwined interests among online live-streaming platforms, online celebrity anchors, and live-streaming users have gradually brought live-streaming platforms into the field of criminal law by using technical support and other means to provide assistance for crimes. As an important bridge and pillar connecting consumers and Internet celebrities, the behavior of live-streaming platforms concealing and shielding Internet celebrities from selling fake and shoddy products and providing protective umbrellas for criminals will not only cause huge property losses to the public but also seriously threaten the stability of economic development and market order. Therefore, how to effectively prevent this phenomenon has become a practical problem that needs to be solved urgently.
As the primary subject of risk control and primary object of supervision of compliance operations, the online live-streaming platform is the key to controlling the chaos of online live broadcasting. In view of this, based on the dual identities of live-streaming platform managers and perpetrators, the revenue cost from the perspective of crime economics is first analyzed to obtain the influencing factors of online live-streaming platform crime decision-making in this paper. Secondly, a four-way evolutionary game model of online celebrities, live streaming platforms, consumers, and regulatory authorities are constructed, mainly discussing the influencing factors and evolutionary drivers of criminal and management decisions of online platforms. Finally, countermeasures to prevent online live-streaming crimes are proposed. The innovation of this article lies in the fact that: (1)Considering the dual identities of live-streaming platforms. (2)The four-way evolutionary game model is constructed, and the participants in the system are included in the evolutionary game model. (3)Using the theory of crime economics, the cost and benefit analysis of the cost and benefit of the live broadcast platform with the sticky characteristics of the “Internet celebrity economy” is carried out for the first time.
The findings of this paper are as follows. First, as a perpetrator, the criminal decision-making of online live-streaming platforms is jointly affected by factors such as illegal income generated by illegal rent-seeking, operating costs, negative reputation value, crackdown strength of government functional departments, criminal punishment intensity, investment cost and credibility of crime control. Second, as a manager, the strategic choice of an online live-streaming platform is closely related to the quality control of online celebrities with goods, the degree of consumer rights protection, and the strategic choice of strict government supervision. When online celebrities and online live-streaming platforms reach illegal rent-seeking cooperation, consumers’ active rights protection becomes the best strategy. It can be seen that the different strategy choices of online celebrities and live streaming platforms are not the optimal decisions of the two, and the choice of the same strategy is the optimal decision of the two. Due to the large number and scale of Internet celebrities, it is a good idea to firmly grasp the first line of defense of platform supervision and re-transform the platform from the role of perpetrators to the role of regulatory managers.
Considering the influence of the dual identities of live-streaming platform managers and criminals on the subjects of live-streaming systems, this study reveals the strategic selection mechanism of each game subject through the four-way evolutionary game, which provides theoretical thinking for solving the problem of online live-streaming crime in the era of Internet celebrity economy. However, in the period of great development of live broadcasting, the decision-making process of each subject will be dominated by subjective internal emotions. So, in further research, the influence of subjective preference factors of game subjects on system evolution and results will be considered, and multi-period game comparisons will be carried out.

In recent years, the platform has gradually derived an ecosystem, developed layout in core business and complementary fields, creating fierce competition for resources in the same field. The value co-creation logic based on consumer experience promotes the deep integration of platforms, complementary enterprises and users, and the interaction between value co-creation subjects helps the platform ecosystem to explore new business value and bring higher quality service and consumption experience to users. In the economic model of consumer Internet platform, value co-creation has become a new paradigm for the platform ecosystem to expand the business landscape.
However, most of the studies have demonstrated the enabling effect of incentive behavior on the realization of value co-creation on the theoretical or institutional level, but have not deeply discussed the mechanism of specific incentive behavior affecting value co-creation from the perspective of the mathematical model. At the same time, the platform research in the traditional bilateral market only regards the platform enterprises, complementary enterprises and users as the supply-demand match, which separates the system integrity of the platform ecosystem and thus fails to reflect the value co-creation effect among ecological subjects. Therefore, it is of great significance to study this scheduling problem.
With the consumer Internet platform ecosystem as the research object, first, we build the value co-creation concept model of the consumer Internet platform ecosystem, clarify the mechanism of value co-creation within the ecosystem. Secondly, based on the conceptual model, we build the three-party collaborative evolution game model of platform enterprises, complementary enterprises and users from the perspective of limited rationality, which reflects the enabling function of platform enterprises in the form of external incentive, systematically analyze the influence of decision-making behavior on value co-creation of platform ecosystem, explore the stability of the local evolution of the system under the strategy equilibrium, and discuss the mechanism of different decision behaviors of the three parties. Finally, based on the leading role of platform enterprises, combined with numerical simulation, we analyze the promotion effect of platform technology incentive, supplier incentive and user incentive behavior on other subjects’ decision-making and value co-creation, explore the path to improving the value of platform ecosystem under the power of platform enterprises, and provide new ideas for the subsequent development of platform economy.
In the tripartite evolutionary game model, three stable strategy combinations and two stable co-creation relations of the platform ecosystem are obtained, and the equilibrium point of the mixed strategy is not the evolutionally stable strategy. The results show that technology incentive, supplier incentive and user incentive can drive platform enterprises, complementary enterprises and users to co-create synergistic value. The user incentive has the greatest effect on the platform ecosystem value co-creation performance, while the supplier incentive has the least effect. The role of technology incentive is mainly manifested in driving users’ willingness to participate in value co-creation, which can significantly improve the speed of value co-creation. The value of user incentive mainly lies in accelerating the participation of platform enterprises in value co-creation.
The main contributions of this paper are as follows: we discuss the influence of the decision behavior on the realization of the value, provide the theoretical and practical value for the strengthening of the cooperation between the ecological network of the platform ecosystem under the flow dividend, and the guiding point for the sustainable development of the platform ecosystem in the fierce competition environment.

As a fundamental industry of the national economy, mining industry is characterized by capital of large amount and for a long period. Supply chain finance is an effective measure to relieve enterprises of capital pressure, because it not only can improve the capital flow of supply chains, but also be widely applied to other sectors such as agricultural, coal, petroleum, and other commodity industries. Since the 21st century, the pricing mechanism of iron ore has shifted from annual agreement pricing to quarterly index one. In recent years, the supply and demand patterns of global mineral resources have also been changing constantly. These phenomena have exacerbated the uncertainty of iron ore prices, making iron ore price fluctuations more frequent. Price fluctuations have always been the main risk factor in commodity trade, certainly including iron ore. The impact of its price fluctuations on upstream and downstream enterprises in the iron ore supply chain should not be underestimated. Through a comprehensive literature review, this study finds that existing studies of iron ore supply chain finance are still rare, especially from the perspective of iron ore price fluctuations. As the world’s largest importer of iron ore, China is highly sensitive to iron ore price fluctuation. In this context, this paper aims to develop a novel advance-payment model by considering iron ore price fluctuation, profoundly analyze the optimal equilibrium decisions of mining enterprises and steel mills, and explore the impact of iron ore price fluctuations on enterprise decision-making and expected revenue. Therefore, this study provides a theoretical foundation for industrial practitioners to make better decisions in an analytical way.
This paper alternatively applies the supply chain finance model, advance-payment mechanism, Stackelberg game and price uncertainty method to the iron ore supply chain. By investigating the properties of yield uncertainty and advance payment, this paper devises a wholesale-price discount coefficient to formulate a two-level Stackelberg game model, in which a steel mill acts as a leader while a capital-constrained mining company represents a follower. This study aims to obtain better equilibrium decisions for both players (i.e., the leader and follower) and evaluate intrinsic characteristics of iron ore price fluctuations. Based on the real-world data, a sensitivity analysis is conducted by considering the mining company’s initial capital and iron ore price fluctuation. This paper uses the MATLAB software to compute the expected revenue and analyze equilibrium decisions in a real-world iron ore supply chain case. The results show that the optimal order quantity of steel mills is about 145 million tons, the optimal planned production quantity of mining enterprises is about 78 million tons, and the overall expected revenue of the iron ore supply chain is about 244 billion yuan. When the iron ore price rises, mining companies intend to increase their planned production quantity and obtain more expected profit. Meanwhile, steel mills keep their optimal order quantity constant to gain more revenue. Another important finding in this case study is that, because of the higher return rate and the more severe capital requirement which mining companies are faced with, steel mills are willing to participate in supply chain finance when the iron ore price is higher. This paper contributes to making a theoretical breakthrough in the area of supply chain finance and provides a real-world application to the iron ore mining industry. For the future research directions, this can be further explored by integrating other key factors (e.g., the results of iron ore production planning and scheduling) influencing the whole iron ore supply chain procedure, to establish a supply-chain-finance risk-warning mechanism for upstream and downstream mining enterprises.

With the development of the economy and consumers’ demand for high-quality products, enterprises have to strengthen their competitiveness through the overall planning of the supply chain. Considering only the traditional single-channel operation mode, we are unable to meet the current market requirements, so the multi-channel distribution choice of products is now the problem of supply chain network planning. Compared with other products, perishables have the perishability. The quality of products will decrease with time, affect consumer satisfaction with the products and increase the cost of the supply chain network. The operational model for improving the competitiveness of perishables enterprises should simultaneously consider the cost and customer satisfaction.
This paper combines the concept of perishables and constructs a perishable supply chain network, consisting of multiple manufacturers, processing factories, distribution markets, and consumer markets. This supply chain network can be applied to different types of tangible perishables. A cost function is constructed based on the multi-channel distribution network, where the total cost of perishables is jointly influenced by factors such as their own sales price, transportation time, transportation volume, and the sensitivity coefficient of perishables themselves. Simultaneously, based on the initial freshness of the product and transportation time, the customer satisfaction function is constructed to achieve the goal of minimizing the cost of the perishable supply chain network and maximizing customer satisfaction. By using linear weighting method, the multi-objective is transformed into a single objective numerical problem to solve the problem of minimum total cost and maximum customer satisfaction for different weights by the decision maker, plan the perishables supply chain network and provide a basis for management decision-making.
Aiming at the deficiencies of sparrow search algorithm(SSA), an improved sparrow search algorithm (ISSA) is proposed. On the basis of SSA, Tent chaotic mapping is introduced to initialize the population, increase the population number and merge the two populations. Then the elite population is obtained by the elite strategies to improve the quality of initial solutions. Introducing an adaptive periodic convergence factor α is to enhance search capability and convergence speed. The update method for the follower and warning position is adjusted to prevent the algorithm from falling into local optimization to a certain extent. Polynomial variation perturbation is introduced to solve the problem of SSA falling into local optimization. In order to verify the improvement effect of ISSA, genetic algorithm(GA), particle swarm optimization(PSO), SSA and test functions of different dimensions are introduced to predict the performance of the algorithm. The optimization results show that the quality, stability, and convergence speed of the algorithm solved by ISSA significantly improve, which is superior to the compared algorithms mentioned above. The use of time complexity analysis proves that the performance improvement of ISSA algorithm is not achieved by sacrificing time.
This paper takes the supply chain network of a certain dairy product as the research object, provides parameter information of the perishable supply chain network based on enterprise operation data, and uses the improved sparrow search algorithm to solve the model. The results indicate that an increase in total cost mainly depends on the increase in perishable costs, and the sensitivity of perishables to time determines the level of perishable costs. By analyzing different decision weights, the results show that there is a balanced compromise solution in the network planning of the perishable supply chain to obtain the optimal network planning scheme with overall satisfaction.
The model in this paper is based on the assumption that there are no transportation delays or other situations. However, in real life, transportation delays and other situations are commonly present in supply chain networks. Therefore, relevant case data can be collected for further verification and optimization research.

Power grid system is the key infrastructure for the development of modern society. The interconnectivity between different lines in a power grid makes the operation of a power system with long-distance transmission more efficient, but it also causes transmission failures. Due to the inherent characteristics and physical mechanism of transmission, the cascading failures result in large-scale power outages. Cascade failures have fast propagation speed and wide range, which leads to huge social and economic losses.
In order to describe the impact of dynamic changes of traffic in the grid system on the grid operation and to analyze more precisely the reliability of different lines at different moments, this paper establishes a multi-state cascade failure Markov model with dynamic failure rates. It sorts out a variety of methods used in the study of cascade failure of power grid system, and gives the topology structure, the calculation method of flow and the initial state of power grid system. Based on the state space of the lines, the sets of lines in different states of the power grid system are defined respectively. The classical two-state model (operating state and fault state) is expanded into a multi-state one including normal operating state, safe fault state and dangerous fault state. A new multi-state cascade failure Markov model is established to analyze the transition probability of lines and the transition probability of the whole system in the power grid system. Combined with the characteristic that the flow is redistributed with the process of cascade failures, the calculation methods of safety failure rate, dynamic dangerous failure rate, line reliability and system reliability are given. This method can more accurately describe the influence of dynamic flow change on the operation of power grid system and analyze the reliability of different lines at different moments. Finally, the reliability analysis process of power grid system is given by the numerical example, and the differences in reliability of the same power grid system under the fixed failure rate and dynamic one is compared, which shows the effectiveness and accuracy of the model.
This paper expands the classical two-state model into a multi-state model including normal operating state, safe fault state and dangerous fault state. Considering the line failure rate with the dynamic change of flow load, the breakthrough failure rate is often approximately the error caused by fixed constant value. This paper establishes a multi-state cascade failure Markov model with dynamic failure rates, the related reliability indexes are analyzed, and the calculation methods of safe failure rate, dangerous failure rate, most vulnerable line index and system reliability are given, so as to analyze the reliability of multi-state cascade failure power grid system with dynamic failure rate more accurately.

The competition between the high-speed rail (HSR) and air transportation has been intensified due to the fast development of HSR network in China. During the period of COVID-19 pandemic, passengers chose diversified modes of travelling, such as private cars, as both high-speed rail and air transportation would face unpredictable disruptions due to the regional outbreak of the pandemic. The choice for passengers could be complex, as it was highly affected by the fast-changing situation of the pandemic. Thus, the dynamics for passengers to choose the means of transportation, including HSR, airplane, and private cars ought to be analyzed. By considering passengers, HSR, airlines as three decision makers (DMs), the reasonable strategies for HSR and airlines can be obtained from the equilibrium reached by the three DMs which make their available options from the strategic perspective. In particular, the behavior of passengers is investigated by extracting the public opinions on the internet, so that the preferences of passengers can be elicited with more credibility. The strategies analyzed based on the preferences of passengers can provide reasonable implication for HSR and airlines to survive the dynamic environment of competition in the pandemic era.
The competition between HSR and airlines at the strategic level is studied using the Graph Model for Conflict Resolution (GMCR), a flexible methodology to solve strategic conflicts. As the input of model, the preferences for three DMs, HSR, airlines, and the passengers ought to be elicited. By considering the airlines in the aviation industry, HSR industry, and the passengers with the majority of public opinions, each of the three groups is considered an individual DM. To increase the reliability of the preferences for passengers, we utilize a professional platform, called Zhongyiyuqing, to collect and analyze the public opinion regarding the experience of taking HSR, airplane, and private cars. Public opinion is used for determining the preference statements expressed by the options controlled by the three DMs connected by logical symbols. The equilibria, as the output of the model, are calculated based on the four solution concepts reflecting different patterns of behavior of the DMs, using the existing decision support system called GMCR II. Moreover, the evolution of the conflict from the initial status of the competition, called the status quo, to the equilibria is also studied, so as to reveal the equilibrium situation of the competition in reality. The equilibria at the situation with stricter control of pandemic are also calculated and compared with those at the usual situation. At this situation, the level of pandemic control is reflected by the online opinion of passengers.
The equilibria are calculated at multiple scenarios with different opinions of passengers as well as the level of pandemic control. As suggested by the equilibrium accessed in the evolution analysis, HSR and airlines would choose to compete with each other and refuse to cooperate. Quick actions are recommended: whichever moves first would gain the upper hand. In the medium-range market where fierce competition takes place, airplane would be more preferred by the passengers. When the pandemic control is stricter, HSR and airlines will be suggested to cooperate so as to counter the flow of passengers to private cars. It can also be indicated that safety is the primary concern for passengers.
Both HSR and the aviation industry can be suggested with meaningful strategic implications. First, pandemic control rather than the ticket price might be the prioritized for both HSR and airlines, as passenger might choose private cars for travelling when the pandemic appears regionally. Second, both HSR and airlines should make corresponding market strategies for attracting passengers with an intention of driving. Besides, cooperation between HSR and airlines is wise for both sides to survive the pandemic and achieve a win-win situation, especially when the pandemic control is tightened. This study extends the theoretical framework of the graph model by using online public opinion to describe the preference of passengers in choosing means of public transportation. The results could facilitate the development of business strategies for both HSR and the civil aviation industry in China based on the behavior of passengers.

Multi-source procurement of parts and components is a common practice in the industry of consumer electronics. As the transparency of the supply chains for consumer products is enhanced partly thanks to the social media’s report and evaluation of the products, consumers’ knowledge of consumer electronics has improved greatly. Consumers not only are concerned about the whole product, but also know about the parts and components and even their suppliers. Though parts and components from different suppliers pass quality standards, they exhibit performance variation and sometimes result in adverse events. This paper aims to study how the knowledge of the multi-source procurement of parts and components will affect consumer’s purchasing psychology and intention.
Based on the Means-end Chain theory, the customer perceived value theory, and the theory of rational inattention, this paper proposes that multi-source procurement of parts and components will affect the three dimensions of customers’ perceived value, i.e., perceived risk, perceived social value, and perceived brand reputation, which in turn affect consumers’ intention to purchase the consumer electronics with such components. A 2 (whether the differences in key components are easy to perceive) by 2 (whether there is social media coverage) experimental design is used to construct five experimental groups. The laptop is selected as the manipulated product and a Likert-type scale is developed to measure the concerned constructs. Subjects are recruited to participate in the experiment and they are evenly and randomly assigned to the five groups. The subjects in each group are given a description of a manipulated laptop and then asked to fill a questionnaire. A total of 230 effective questionnaires are collected. The reliability and validity of the scale are established by Cronbach’s α and the factor analysis. A structural equation model is then established to verify the proposed model. A single-factor analysis of variance is used to test the difference of the concerned constructs between the multi-source and the single-source procurement situations.
Model fit indicators including CMIN/Df, CIF, NFI, TLI, and PCFI are well above their respective thresholds, indicating a good fit of the proposed model. The statistical analyses and tests generate the following findings. First, compared with single procurement of parts and components, consumers have higher perceived risks, lower perceived social value, and lower perceived brand reputation, which further reduce consumers’ intention to purchase the electronics with multi-source procured components. Second, multi-source procurement of the parts and components that can be easily perceived will increase consumers’ perceived risks, reduce perceived social value and perceived brand reputation. These negative effects on consumers’ psychological states will further weaken consumers’ willingness to purchase the products. Moreover, the social media coverage of differences in the multi-source procured parts and components can also negatively affect the three dimensions of customers’ perceived value, which further reduce consumers’ willingness to purchase such electronics. These results reveal the adverse effects of multi-source procurement of key components on consumers’ purchasing psychology and intention and give suggestions for consumer electronics manufacturers to make prudent multi-source procurement decisions. First, manufacturers should consider consumers’ responses to the multi-source procurement of components in addition to operations considerations. Second, when using multi-source procurement of components, manufacturers had better include the performance indicators of the components that affect consumers’ actual usage experience into the supplier’s evaluation system in addition to technical specifications. Moreover, it is beneficial for manufactures to actively publish the procurement sources of key components rather than wait for social media to report. Meanwhile, manufacturers had better timely and accurately express their commitment to the consistent usage experience for the products with multi-source procured components, reducing consumers’ concerns.
Consumers’ buying decisions are usually driven by complex psychological states. This research makes initial efforts in studying the effects of multi-source procurement of components on psychological states. Future research can investigate how this source practice will affect more consumers’ psychological states, such as emotion and trust. Moreover, future research can consider more moderators, such as consumers’ knowledge of the electronics and study how those moderators modulate the effects of the multi-source procurement of components.

For years of reform and opening up, China has made remarkable achievements in economic development. However, for a long time, the extensive economic growth mode has led to environmental degradation, resource depletion, unbalanced industrial structure, low development efficiency and insufficient technological innovation, and other governance problems, which have seriously hindered the development of green economy and become an important constraint on the realization of China’s sustainable development goals. As the carrier of the real economy, the development mode of “pollution first and then treatment” has been unable to continue for a long time, and the green transformation of industry has become the key to making a change. As an important part of the green financial system, green credit has become the core driving force to promote the green transformation of industry by guiding the flow of funds through the implementation of differentiated credit rationing strategies and limiting the development of “high pollution and high energy consumption” industries. This paper discusses the role and value of green credit in promoting industrial green development from the perspective of spatial spillover, which helps to provide theoretical and practical support for industrial green transformation. Therefore, it is particularly important to study whether green credit can play a given role and what are the influencing factors.
Guided by practical problems, this paper clarifies the impact mechanism of green credit on industrial green development on the basis of considering the spatial spillover effect and threshold effect. First, on the basis of consulting and sorting out relevant literature and policy reports, a more complete industrial green development measurement system has been built from a total of 18 indicators from five dimensions: structural optimization, energy conservation and emission reduction, pollution control, innovation and development, and economic growth, providing effective academic support for green credit to help industrial green transformation. Second, it clarifies the spatial spillover and threshold effects of green credit on the industrial green development from a non-linear perspective. With the in-depth development of regional integration in China, the correlation between regions also increases. The impact of spillover effects on local development is becoming more and more significant. Therefore, in order to explore the nonlinear driving effect of green credit on the industrial green development, taking the industrial green development as the research object, it uses the spatial Durbin model and threshold model to study the spatial spillover effect and nonlinear effect of green credit, and analyzes the impact of green credit on the industrial green development in many aspects.
The main conclusions of this paper are as follows: firstly, on the whole, the development of green credit in China has strongly promoted the green development of industry in the region, and has a significant spatial spillover effect. Secondly, according to the results of the threshold model, there is a double threshold effect of green credit on industrial green development based on the level of economic development, and the promotion effect of green credit on the industrial green development is a marginal increasing nonlinear law. When green credit itself is used as the threshold variable, it is found that there is a significant threshold effect, and the effect of green credit on the industrial green development has an optimal interval, that is, theeffect will be the most significant when it is within the threshold range. Finally, based on the above analysis results, policy suggestions are put forward, including improving the regional coordination mechanism, narrowing the regional gap, improving the construction of the green financial system, and increasing government subsidies and regulation and control.

Listed companies are important to capital market. Improving their quality is the basis for ensuring high-quality economic development. Financial fraud will damage the interests of investors, hence, how to accurately identify financial fraud has become an urgent problem to be sovled.
To represent the true financial situation of listed companies, we designed a three-dimensional integrated index system that contains financial indexes, non-financial indexes and quality indexes. Quality indicators can reflect corporate performance and corporate governance, which is an effective supplement to financial indicators and non-financial indicators. To accurately identify financial fraud companies, this study constructs a financial fraud identification model based on residual neural network with a weight distribution network added at the front of the network. This paper selects listed companies in CAMAR database from 1998 to 2022 as data samples. To ensure the stability of the experimental results, the 10-fold cross-validation technique is used to randomly divide the data set for 10 times according to the above proportion, and the average performance of the 10 test results is taken as the final performance of the model. This study adopts the accuracy rate as the performance evaluation index of the model.
To comprehensively analyze the advantages of residual neural network over traditional machine learning methods and ordinary neural network models, two comparison methods are designed in this study, namely “PCA-SVM” and “PCA-BPNN”. Specifically, the popular principal component analysis (PCA) technology is firstly utilized to perform feature transformation on the three-dimensional integrated index system to eliminate the collinearity of the indexes, the obtained principal components are sorted in descending order according to their contribution degree, and then the principal components are selected in turn as the feature components after dimension reduction, until the cumulative contribution degree of the selected feature components is not less than 80%, finally, PCA-SVM and PCA-BPNN are obtained by taking the selected feature components as the input of support vector machine and BP neural network respectively. The experiments show that the performance of the residual neural network model and PCA-BPNN are significantly higher than PCA-SVM (about 12.9%), which shows that the feature transformation and feature modeling capabilities of neural networks are better than traditional machine learning methods. In addition, the accuracy of the model designed in this study is 8.3% higher than that of PCA-BPNN, which indicates that the model based on residual neural network can fully mine and utilize linear or nonlinear information contained in multi-dimensional indexes than ordinary neural network-based model, and has stronger a practical application potential. To analyze the impact of different network structures on the accuracy of final financial fraud identification, this study has designed six models with different network structures. The experimental results show that the network structure used in this study is significantly better than that of the comparison group, and the final accuracy of financial fraud identification reaches 89.3%, showing a strong practical application value.

On February 2, 2021, the State Council, the Director General of Customs and the State Administration of Taxation adopted a new duty-free policy for outlying islands in Hainan Free Trade Port, allowing duty-free retailers to provide local consumers in Hainan Province with the service of “return to the island for collection”. Before the implementation of the new policy, local customers could pick up duty-free goods at airports, ports and other island pickup points with their personal information after purchasing them from Duty Free shops. This process of picking up goods was relatively simple, but the inconvenience cost of carrying duty-free products to and from Hainan Island reduced consumers’ willingness to buy duty-free products. Consequently, in order to improve consumers’ shopping experience, more and more duty-free retailers provide the service of “return to the island for collection” to consumers through different sales channels. In practice, retailers like Hainan Tourism Duty Free shop offer this service for both online and offline channels, while China Duty Free Group and Global Premium Duty-Free Plaza (Haikou) only offer it for its offline channel. After the implementation of the new policy, local consumers choose to pick up duty-free goods at the point of return to the island when they purchase from offline or online channels. The duty-free retailers then package the products at the point of return for consumers to pick up after they return to the island. As a result of providing this service, the retailer reduces the inconvenience cost associated with transferring duty-free goods to and from Hainan Island for consumers, while at the same time it gives consumers a new way to shop, improving their shopping experience. There is, however, a new challenge for duty-free retailers in channel of operation strategies, as they will have to bear the additional hassle costs involved in retailers’ pick-up points, such as product accumulation costs or additional labor handling costs. Our research questions specifically concern whether duty-free retailers provide consumers with “return to the island for collection” services? If so, what is the most beneficial sales channel for retailers to provide this service? This study is new and interesting, which contributes to expanding the duty-free retailing and omnichannel retailing research. And the conclusions of this study can provide a reference for the duty-free retailer to implement new duty-free policies.
The study focuses on the duty-free retailer with an online and an offline channel based on new duty-free policies for outlying islands at Hainan Free Trade Port. Further, we examine consumers’ buying behavior, duty-free retailers’ optimal pricing decisions, and retailer channel operations. First, using the consumer utility theory, we build game-theoretical models and obtain the optimal strategies under three different channel strategies: the duty-free retailer does not provide “return to the island for collection” service (RN mode), the duty-free retailer only provides “return to the island for collection” for the offline channel (RS mode), and the duty-free retailer offers “return to the island for collection” service for both online and offline channels (RSO mode). Second, we explore the impacts of offering “return to the island for collection” on duty-free retailers by comparing their profits before and after providing this service. Then, we get the optimal channel operation strategy for the duty-free retailer by comparing the profits of the retailers who only provide “return to the island for collection” for offline channel and offer “return to the island for collection” service for both online and offline channels. Finally, we extend the model under which the retailer only provides “return to the island for collection” for the online channel and study the duty-free retailer’s optimal pricing decisions.
First, when the retailer’s additional profit or the cost of consumers carrying duty-free goods is low or moderate, but the hassle cost of retailers’ pick-up point is large, it will not be beneficial to the duty-free retailer to provide “return to the island for collection” service. Second, when the retailer’s additional profit is high or the hassle cost of retailers’ pick-up point is moderate, it will be more profitable to provide “return to the island forcollection” service only through the offline channel. Third, when the cost of consumers carrying duty-free goods is large or the hassle cost of retailers’ pick-up point is relatively low, it will be more profitable to provide “return to the island for collection” service through both online and offline channels. In summary, our results can provide a theoretical basis and decision-making reference for duty-free retailers’ channel operation strategies. However, the first limitation of this paper is that we only consider the channel operation strategy of a single duty-free retailer under the “return to the island for collection” policy. And the second limitation of this paper is that we only investigate the effects of “return to the island for collection” policy on duty-free retailer’s optimal decisions. The further research can be carried out on the operational decision of multiple duty-free retailers under the “return to the island for collection” policy or other new policies.

In the context of economic globalization, the business and financing fields of enterprises are expanding. In order to enhance the competitiveness of enterprises and achieve long-term development, selective implementation of several projects has become an effective competition mechanism. Therefore, on the premise of meeting various constraints, it is very important to select the project portfolio that can make the enterprise obtain the maximum benefit from many alternative projects. This kind of problem is called project portfolio selection problem.
In the field of project portfolio selection, because many projects are involved, the precedence relationship between projects has attracted the attention of relevant scholars. However, in the existing research, the precedence relationship strictly requires that the project should be implemented only after the predecessor project is completed, but ignores the fact that the follow-up project can be implemented after part of the predecessor project is completed. In addition, in the field of portfolio selection, when there are multiple predecessor projects, in addition to the existing two predecessor project completion quantity constraints, there will also be cases where only a certain number of predecessor projects need to be completed. Therefore, it is necessary to consider this situation in the field of project portfolio selection to realize more scientific and reasonable decision-making.
In view of the above problems, this paper makes further research on the basis of the existing precedence relationship constraints. Firstly, this paper perfects the constraints of precedence relationship, and considers that the completion of a certain number of predecessor projects can complete subsequent projects. Secondly, based on the aforementioned constraints, a new project portfolio selection model is developed. Finally, on the basis of both data from existing literature and simulation results, the model is solved by the optimization solver CPLEX. Further, the sensitivity analysis of the parameters involved in the constraints is carried out, and the relevant conclusions are drawn. The specific contents are mainly divided into the following three parts.
The first part introduces the sets, parameters and variables required in the modeling process, so as to prepare for the subsequent modeling process. This step is essential as it establishes a clear framework, ensuring that the subsequent modeling steps can be executed systematically and efficiently.
The second part focuses on depicting the perfect precedence relationship constraints, and the fact that after a certain number of predecessor project complete, follow-up projects can proceed. On this basis, combined with the constraints in the traditional portfolio selection problem, the portfolio selection problem affected by precedence relationship in the interruptible situation is established.
In the third part, an example is given to verify the effectiveness and practicability of the MCPRD model proposed in this paper. Firstly, the PPSP-1 model of “all predecessor projects must be 100% completed”, and the PPSP-2 model of “all predecessor project precedence relationship parameters and the number of predecessor projects must be completed” are proposed, and based on data from existing literature and simulation results, and the optimization solver CPLEX, the results of comparison between different models verify the practicability and effectiveness of this model through different scale examples.
To sum up, this paper supplements and improves the existing precedence relationship constraints in the field of portfolio selection, and constructs a portfolio selection model affected by precedence relationship in interruptible situations. The results show that: 1)considering the fact that “a certain proportion of predecessor projects can be completed to implement follow-up projects” and “when a certain number of predecessor projects are completed, follow-up projects can proceed” in the problem of project portfolio selection, the resulting decision-making scheme will be due to the traditional model. 2)Before formulating the implementation plan, enterprise decision-makers should carefully dig out the precedence relationship between projects. When there are multiple predecessor projects, we should also pay attention to the coincidence between multiple predecessor projects, so as to avoid the strict sequence of front and rear implementation, which makes high-yield projects unable to be implemented and affects the income of the enterprise. 3) The impact of changes in the number of predecessor projects to be completed on income is much higher than that of the precedence relationship parameters. Decision makers should reasonably arrange resources, reduce the required number of precedence projects as far as possible, and improve the income of project portfolio selection.
There are still several issues that require further exploration in this paper. For instance, the risk of losses resulting from initiating follow-up projects after only a certain percentage of predecessor projects are completed has not been fully addressed, and risk transfer mechanism between projects under the precedence relationship has not been fully explored. These things will be key focuses for our research group’s future work. In addition, we are indebted to the North China Electric Power University Library for providing detailed reference for our research, which has greatly facilitated our study and enriched the depth of our analysis.

Agricultural industrialization alliances, formed by a diverse array of agricultural operators through contractual agreements, have emerged as a significant organizational model driving agricultural development in the new era. However, in practice, the phenomenon of alienation within these agricultural alliances is frequently observed. This includes issues such as the club-like nature of member organizations, the de-familization of primary operators, and excessive scaling that effectively marginalizes small farmers. Such trends severely deviate from the central government’s policy vision for establishing a modern agricultural management system and achieving strategic goals related to rural revitalization. The underlying cause of these challenges lies in the absence of an effective benefit-sharing mechanism among stakeholders within the agricultural alliance that fosters continuous improvement in agricultural performance. Consequently, this study aims to explore the design of a rational performance incentive mechanism to promote cooperative participating behavior among farmers in these alliances. Establishing a novel cooperation framework that facilitates interest sharing and mutual benefits between primary operators and farmers is crucial for stimulating endogenous motivation toward autonomous alliances among various stakeholders engaged in agricultural land management.
To maximize the performance output of agricultural industrialization alliances and optimize the distribution of benefits among agricultural land operators, this study utilizes agency theory to examine the incentive-related issues associated with the performance of an agricultural industrialization alliance comprising a new type of agricultural operator and multiple farmers. The production efforts exerted by farmers are classified into self-interested efforts and altruistic (cooperative) efforts. A performance incentive mechanism model based on total alliance performance output (UA) is developed alongside an incentive model focused on individual farmer performance (NA). These models are subsequently solved, a sensitivity analysis is made and numerical simulations to assess the results are derived from these models.
The results show that: (1)The effects of the UA and NA incentive mechanisms will be influenced only when the performance threshold set by agricultural regulatory authorities exceeds the threshold determined by equilibrium performance output, allowing the government to guide the behavior of agricultural industrialization alliances through this “target incentive” directive. (2)In low reserve performance scenarios, the two types of farmers’ effort levels, the personal performance incentive coefficient, the alliance performance incentive coefficient, and the total alliance performance output are positively correlated with the effort-performance conversion rate and negatively correlated with the risk aversion factor, while in high reserve performance scenarios, the opposite relationships are observed. (3)Compared to the NA incentive mechanism, the UA incentive mechanism effectively enhances farmers’ enthusiasm for cooperative production and fosters strong cohesion within the alliance. This is conducive to achieving a higher balanced performance output, thereby enabling new agricultural business entities to realize positive economic value benefits. Simultaneously, it stimulates internal motivation among diverse interests related to agricultural land, promoting performance improvement, benefit-sharing, and coordinated development within the agricultural industrialization alliance.
Based on the aforementioned conclusions, the managerial implications are as follows: (1)New entities should be actively guided to establish a comprehensive performance incentive mechanism grounded in a “cooperative” agricultural alliance, thereby fostering collaborative enthusiasm among participating farmers. (2)The government’s “target incentive” directive serves as a guiding framework for influencing the behavior of agricultural alliances, promoting sustainable and high-quality agricultural development while stabilizing alliance performance outputs and mitigating uncertainties faced by farmers. (3)Utilizing performance metrics established by regulatory authorities, we can evaluate the number of farmers engaged in the “collaborative” agricultural alliance to achieve optimal total performance output. This strategy aims to facilitate new entities in realizing positive economic value gains.
Future research may focus on analyzing farmer heterogeneity, examining incentive mechanisms when the new agricultural operator and farmers have different risk preferences, and empirically testing the conclusions drawn from the agency theory framework.

COVID-19 has had a profound impact on the global economy and financial activities, driving a surge in consumer demand for online shopping and a sharp increase in business volume for e-commerce platforms. Achieving accurate product recommendations based on consumer demand has become a hot topic of research. Online product reviews are crucial for consumers to assess product quality and merchant credibility, directly influencing purchasing decisions. Therefore, leveraging online review information for intelligent product recommendation is essential. Compared to user ratings, online text reviews can more precisely express various emotional product preferences, often coming in uncertain emotional descriptions such as ‘very like it’, ‘like it’, ‘dislike it’, etc. But transforming the data into fuzzy sets presents complexity and challenges.
Existing research lacks consideration for cases where a single criterion contains multiple emotional preferences, leading to information loss during comment processing and transformation. In real-world decision-making scenarios, customers often exhibit bounded rationality, necessitating further research on product recommendation for loss-averse customers based on rich emotional preference online text reviews. Presently, limited research considers customer loss aversion psychology, primarily utilizing the prospect theory and TODIM method to characterize bounded rational behavior. However, traditional TODIM methods encounter application paradoxes and complexities, resulting in biased purchase decisions. Therefore, a new product recommendation model is proposed based on online text reviews by integrating belief structures, uncertain probabilistic linguistic term sets (UPLTSs), and the generalized TODIM method to address multiple emotional preferences and characterize customer loss aversion psychology effectively.
Initially, a 5-level linguistic term is employed to convert user reviews, considering the nature of online reviews. Belief structures are introduced to fully leverage their powerful uncertain information processing capabilities in transforming multiple emotional preferences in text reviews. Next, ‘benefit type’ and ‘cost type’ criteria in product reviews are normalized before calculating criterion weights based on the similarity between evidence bodies. The higher the similarity, the greater the criterion weight assignment. Subsequently, belief structures, UPLTSs, and the generalized TODIM method are integrated to devise a novel product recommendation model suitable for loss-averse customers with multiple emotional preferences. To efficiently address uncertain probability linguistic issues, on one hand, the reliability function and plausibility function of belief structures are used to transform evaluations, resulting in an uncertain probability linguistic evaluation matrix. On the other hand, to mitigate the paradoxes and complexities associated with the traditional TODIM method in application, the generalized TODIM method is extended to an uncertain probability linguistic environment, effectively capturing customers’ loss aversion psychology as observed in real decision-making scenarios. This streamlines the computational processes and broadens the application scope.
To demonstrate the effectiveness of the recommendation model, a case study on mobile phone recommendations is conducted. Four phones are selected based on customer requirements, recommending the highest-rated phone. To validate the robustness and effectiveness of the proposed model, a sensitivity analysis is conducted using loss aversion coefficients, and a comparison is made with existing researches. The results show that changes in the degree of risk aversion do not affect the product recommendation order, indicating insensitivity of the recommendation results to changes in the level of risk aversion, thereby demonstrating the robustness of the proposed recommendation model. Through the comparative analysis, it is evident that in a product recommendation model based on text reviews, handling and transforming multiple emotional preferences, and considering customers’ bounded rational behavior are extremely necessary, resulting in recommendation outcomes that better align with human innate thinking habits, reflecting customers’ needs more authentically, and holding significant theoretical and practical value for actual product recommendation.
This study focuses on personalized product recommendation for specific needs. Future research will consider customer groups without specific needs and apply the proposed recommendation model to other fields. Through this extension, deeper insights will be provided for website optimization management strategies and a new research perspective will be offered for personalized recommendation problems in different domains.

The supply and quality of fresh agricultural products are related to the national economy and people’s livelihood. Different from other products, the perishable nature of fresh products inevitably leads to the loss of quantity and quality in the circulation process, which has caused a lot of losses to enterprises for a long time. As an important means to maintain freshness and reduce loss, the cold chain has become an obstacle in the development process due to its high construction cost, which makes it difficult for many fresh produce supply enterprises to afford this expense. In order to solve the problem of capital constraints, in reality, companies can get out of the dilemma by means of financing, and different financing methods will produce different results. In order to determine the optimal financing method and profit of suppliers, this paper constructs a two-level fresh agricultural product supply chain model composed of suppliers and retailers to solve the problem of insufficient funds for the construction of suppliers’ cold chain. The two sides make decisions according to the Stackelberg game, and introduce the factors of suppliers’ fresh preservation level and freshness loss. Meanwhile, three financing modes such as retailer’s advance payment, bank loan financing and mixed financing of bank loan & equity are considered for suppliers, and the optimal decision and profit of supply chain members are obtained by using the optimization theory and method, and the best financing mode and applicable conditions for the cold chain construction of suppliers are also analyzed. Combined with official data from the People’s Bank of China and publicly available research data from scholars in the field, an arithmetic analysis is used to verify the effects of the retailer’s advance payment ratio, equity dividend ratio, relative benefit ratio, and product perishability on suppliers’ cold chain preservation decisions and profits, and the validity and accuracy of the research findings are confirmed. The study finds that in the case without financing, the initial capital will have an impact on the decision-making and profits of supply chain enterprises, and the higher the initial capital level, the better the cold chain preservation level and the improvement of supply chain enterprises’ profits, and the more attractive the cooperation of downstream retailers. Under the retailer’s advance payment mode, when the initial capital level is higher than a certain threshold, the supplier’s operation decision and the profit of supply chain members can reach the optimal level, which can effectively solve the dilemma of supplier’s capital constraint. Otherwise, the supplier cannot reach the optimal decision and profit, and still has not got rid of the dilemma of capital constraint. Suppliers’ investment in cold chain preservation will gradually increase with an increase in product perishability. When the advance payment ratio is low, the cold chain preservation level will be the best under the retailer’s advance payment, but when the advance payment ratio is high, the adoption of the mixed financing mode will be more conducive to the improvement of cold chain preservation level. The supplier’s financing strategy is affected by the retailer’s advance payment ratio, equity dividend ratio, relative profit ratio and product perishability. When the proportion of prepayment is low, the profit of retailers under advance payment will be always higher than that of other financing situations. Conversely, bank loans and mixed financing are more favorable; the bank loan approach will be more favorable when the relative benefit ratio is larger and the equity dividend ratio is higher. The research conclusions of this paper also give many inspirations to supply chain related enterprises. In order to improve the quality of fresh products and increase revenue, capital-constrained suppliers should first consider the retailer’s advance payment ratio, equity dividend ratio, relative profit ratio and product perishability, in combination with the amount of their initial capital when raising funds for cold chain preservation construction, so as to develop the optimal financing method. At the same time, the findings of the study can also provide theoretical basis for the cold chain construction of fresh produce enterprises, which can help promote the high-quality development of the supply side of fresh produce and further strengthen the construction of the “Vegetable Basket Project”. In the future, we can also consider the uncertainty of demand, so the research on this issue will be more meaningful, and we can further take into account the information asymmetry and enterprise risk preference to broaden the research on the intersection of supply chain finance and fresh agricultural products supply chain.

As the Chinese stock market matures and market efficiency continues to improve, the application of statistical arbitrage strategies is growing increasingly. The narrowing profit margin for single-stock pairing strategies poses a challenge to the optimization of traditional pair trading strategies. As China gradually relaxes restrictions on margin trading and introduces short-selling mechanisms, the operational efficiency of the capital market has been significantly enhanced. The rapid development of margin trading and the continuous expansion of target stocks have provided more arbitrage opportunities, indicating substantial potential and application for pair trading strategies in China.
This paper introduces a multi-stock pair trading optimization strategy-the MultiPT optimization strategy, which establishes a flexible framework for pair trading stock selection and expands the research scope of pair trading methods. The proposed optimization algorithm overcomes the limitation of traditional pair trading strategies, which require an equal number of matched pairs, enhancing the flexibility and practicality of stock selection. Based on the minimum distance method and cointegration test idea, the study seeks stock pairs with the minimum price distance, considering non-zero stock weight constraints. It tracks dynamic stock price differentials through cointegration tests to capture arbitrage opportunities in long-term equilibrium relationships. Given the non-linear nature and constraints of the optimization model, the Sequential Quadratic Programming (SQP) algorithm is used effectively to solve the model, finding multiple locally optimal solutions and determining the optimal weight distribution between two sets of stocks to construct the best pair combinations.
The study selects the constituent stocks of the CSI 300 Index and the Wind Industry Index from the Chinese A-share market as back-testing samples, spanning from January 1, 2015, to December 30, 2020. The back-testing results show that, compared to traditional GGR and cointegration strategies, the multi-stock pairs trading model achieved excess returns of at least 15.17% and 14.97% in the CSI 300 constituent stocks and Wind industry samples, demonstrating significant advantages in multi-stock pairing. The study also finds that the MultiPT optimization strategy yields the highest returns with two stocks in the pairs, with strong risk hedging capabilities. However, as the number of stocks in the pairs increases, the stability of the strategy decreases; when the number of stocks reaches a preset limit, despite reduced returns, higher strategy stability will be maintained. This indicates that under certain conditions, increasing stock pairs can help diversify risk, but may reduce returns. Notably, the GGR, cointegration, and MultiPT optimization strategies all perform more prominently during market downturns. Overall, the multi-stock pair trading strategy shows robust performance in various market conditions and samples, consistently outperforming the single pairing algorithms in GGR and cointegration strategies.
The research results confirm the feasibility and effectiveness of the multi-stock pair trading strategy, providing a model, algorithm guidance and empirical validation for pair trading strategies in the Chinese stock market. This study offers detailed evidence for investors to understand and implement pair trading strategies in various market situations, providing theoretical support and practical guidance for exploring stock market vitality and promoting high-quality development in China’s financial market. Future research will design selection rules based on investment preferences and constraints, employing a broader and more diversified dataset to verify the robustness and effectiveness of the multi-stock pair optimization algorithm.

The returns and risks of assets in the securities market are uncertain. The core problem faced with by investors is how to optimize the allocation of wealth in an uncertain environment so that the risks are minimized or the returns are maximized. To achieve this goal, the artificial intelligence and optimization methods are used by the investors studying portfolio selection problems, which provide investors with new theoretical frameworks and investment strategies. And numerous widely accepted empirical researches suggest that stock prices and returns are predictable to some extent. In this case, it is necessary to observe the change and volatility of financial data over a long time in the past so as to make a good preparation for future trend forecast and investment decisions. Up to now, existing studies have mainly focused on: (1)Statistical methods. They aim at prediction by analyzing the past price characteristics, such as linear regression, autoregressive conditional heteroscedasticity, autoregressive integrated moving average, and generalized autoregressive conditional heteroscedasticity model. And (2)machine learning methods. They are K-Nearest Neighbor, support vector regression, eXtreme Gradient Boosting, deep neural network, long-short term memory, convolutional neural networks, etc. Some comparative studies emphasize that machine learning has stronger ability to deal with non-linear and non-stationary problems than statistical models.
In this paper, a portfolio selection model is proposed using machine learning with hyperparameter optimization for the stock prediction and Mean-variance model for portfolio selection. To be specific, two stages are involved in this model: the stock prediction and portfolio optimization. In the first stage, a hybrid model combining XGBoost with a modified firefly algorithm based on specific probability (pFA) is proposed to predict stock prices for the next period, and compares its predictive ability with XGBoost, LSTM and SVR algorithms. The pFA is developed to optimize the hyperparameters of the XGBoost. In the second stage, the stock selection (stocks with higher predicted returns are selected) and asset allocation (spreading funds across selected stocks). Considering the constraints such as transaction costs and threshold constraints, the mean-variance model and equally weighted model are employed for the portfolio selection. The MV model aims to make a trade-off between maximizing returns and minimizing risks, which is expressed by a typical multi-objective optimization formula, and introduce the risk aversion coefficient to change the multi-objective formula into the mono-objective formulation.
At the same time, using China Securities 300 Index component stocks as study sample, we give a numerical example to demonstrate the designed algorithm’s performance and the proposed model’s application. We specifically and randomly select 48 stocks in the CSI300 as candidate assets, large enough for individual investors to choose stocks before forming portfolios. The sample interval is from January 2013 to December 2021. The history data is divided into six periods, every period containing four-year data divided into a training set and a testing set as a ratio of 8∶2. The training set is used to train the model and adjust the hyperparameters to get a good generalization, and the test set to evaluate the performance of the final model.
In this paper we select 19 indicators as the input of the stock prediction, and the MV model is used to conduct the portfolio’s asset allocation based on the selected high-quality asset. We use the pivoting algorithm to solve the MV model without short sales and to obtain the optimal portfolio strategy. In addition to the MV model, an equal-weight portfolio is also studied. To investigate the accuracy of stock prediction methods, four indexes, namely mean square error, root mean square error, mean absolute error, and hit ratio are used. From the experimental results, we have several important findings: (1)to improve the FA’s optimization ability, the pFA is developed, and after comparing the pFA with FA, PSO, AFSA, GA, and DE, the advantage of pFA is verified by a set of unimodal and multimodal test functions; (2)the empirical results demonstrate that the pFAXGBoost+MV model achieves better results than its counterparts and the market index in terms of return and return-risk metrics.
Considering realistic constraints, a portfolio selection model is proposed using machine learning with hyperparameter optimization for the stock prediction and Mean-variance model for the portfolio selection. And it is a convex quadratic programming problem with equality and linear inequality constraints, which is solved by a novel improved pivoting algorithm. On the one hand, it enriches the research into the modern financial decision-making theory and provides a new idea based on machine learning for the stock prediction. On the other hand, it helps investors adjust their investment strategies in the quantitative investment, and enhances the ability of individual and institutional investors to adapt to the investment environment. However, there are some limitations to this study. This paper can further be improved and extended from the following aspects: (1)Semi-variance, VaR, and skewness can be used for portfolio selection. (2)Cardinal constraint and minimum trading volume can be considered.

In the context of a dynamic and complex digital economy, the healthy development of digital finance, as the blood of the digital economy, is very important to the whole economic system. Digital finance has gradually penetrated every field of social economy, changed people’s mode of production, lifestyle, and consumption behavior, and produced new forms of financial business and innovative service methods. These digital financial innovations have become the source of the development of the digital financial industry, and their evolutionary path and results also affect the development process and direction of the entire financial industry. However, investment behavior will occur only when investors realize that digital financial innovation can bring wealth appreciation and improve financial efficiency. Therefore, investors’ perception of digital financial innovation is the main driving force for digital financial innovation. Based on the above background, from the perspective of investors’ perceived value, this paper uses the method of the evolutionary game to build an evolutionary game model composed of digital financial platforms and investors and analyzes the evolutionary path and evolution law of digital financial platform’s innovation behavior and investors’ investment behavior.
Firstly, considering the value perception of investors, this paper first adopts the method of the evolutionary game to construct an evolutionary game model composed of a digital financial platform group and an investor group. On this basis, the paper analyzes the evolutionary results, evolutionary paths, and evolutionary rules of the innovation behavior of digital financial platforms and investors’ investment behavior. Secondly, taking the quarterly panel data of Tianhong Yu’ebao Fund from 2014 to the third quarter of 2020 as samples, relevant variables are selected from the four aspects of scale share, platform innovation, investors’ financial literacy, and traditional finance to describe the evolution of Yu’ebao market and its main influencing factors. Thirdly, the system dynamics model of Yu’ebao market evolution is constructed, and the conclusions of the model are verified through data analysis to explore the impact of investors’ value perception and other factors on digital financial innovation.
The results show that: First, there are five equilibrium points in the evolutionary game between investors and digital platforms, but only the equilibrium points E₁(0,0), E₂(1,0) and E₄(1,1) are evolutionary stable points under certain conditions. Under different conditions, the evolution path of the system from an unstable point to a stable point is different. Second, under the condition of meeting the stable point, no matter how the initial entry rate changes, the digital financial system always tends to evolve towards the stable point. Moreover, with an increase in the initial entry rate, the evolution speed of digital financial systems towards this point is accelerated. The higher the value perception coefficient of investors, the higher the probability of investors choosing “platform investment” and the more inclined digital financial platforms is to choose the “innovation” strategy. Third, under the influence of platform innovation and investors’ value perception, the net asset quantity of Yu’ebao market shows an evolutionary trend first rising and then declining. The possible evolutionary stability points in the evolutionary game have been reflected in the development of Yu’ebao market. Investors’ value perception has a positive impact on the benign evolution of Yu’ebao market. Therefore, to promote the development of digital finance, we should attach importance to the role of investors’ value perception. Investors should not only actively share their experience in financial products and services and give feedback to the platform about their opinions or suggestions, but also strengthen their financial knowledge, especially the accumulation of digital financial knowledge, and improve their cognitive ability of innovative products or services of digital financial platforms. Digital financial platforms should pay attention to the construction of their functions, improve their innovation ability, and maximize the launch of digital financial services and products that meet the needs of investors.

With the advent of mobile internet and the evolution of intelligent transportation systems, online car-hailing services have become increasingly popular as a primary mode of transportation. These platforms, such as Didi Chuxing and Uber, have revolutionized the way people travel by reducing information asymmetry and enhancing the efficiency of resource allocation. They achieve this by effectively matching the supply and demand between passengers and drivers. However, as the frequency of usage has grown, the problems with unreasonable vehicle scheduling and mismatch between supply and demand arise, including extended waiting times for passengers and high empty load rates for drivers. To address these issues, numerous scholars have proposed various deep learning methods to predict car-hailing demand more accurately. Despite these efforts, current forecasting methods still fall short in fully exploiting the features of time series data.
In light of this gap, our paper introduces a novel hybrid prediction model known as Convolutional Attention mechanism Bi-directional Long Short-Term Memory (CNN-ATTBiLSTM). This model innovatively captures time series features from both macro and micro perspectives, utilizing both forward and reverse directions to enhance the accuracy of car-hailing demand forecasts. This approach aims to mitigate the problems with suboptimal vehicle scheduling and the imbalance between supply and demand.
We validate our model using a dataset of car-hailing orders from HaiKou in June 2017, provided by Didi Chuxing’s Gaia Plan. The process begins with data cleansing to remove invalid entries, followed by resampling the data at ten-minute intervals. We then employ the K-means algorithm to identify passenger hot spots, which are used to define the unit regions for analysis. The model, CNN-ATTBiLSTM, is fed with external factors such as weather conditions and the demand for unit areas and time slots to forecast demand for the next time slot. To demonstrate the superiority of our model, we compare it with LSTM and CNN-LSTM models using evaluation metrics like Coefficient of Determination (R²), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Symmetric Mean Absolute Percentage Error (SMAPE). Additionally, we visually compare the forecasting and actual demand curves to analyze the models’ fitting degrees during peak and off-peak times.
The experimental results indicate that CNN-ATTBiLSTM outperforms both LSTM and CNN-LSTM across all evaluation metrics. Our visual analysis reveals that while the LSTM and CNN-LSTM models exhibit a degree of lag in their predictions, our CNN-ATTBiLSTM model more effectively captures the dynamics of time series data during both peak and off-peak periods. Theoretically, the CNN component extracts local features, the Attention mechanism assigns varying weights to these features to model their global relationships, and the BiLSTM component provides robust temporal sequence feature extraction capabilities. Collectively, these elements enable the CNN-ATTBiLSTM to model time series features comprehensively, enhancing demand prediction accuracy and providing valuable insights for car-hailing platforms to devise policies for efficient vehicle scheduling.
However, there are still some improvements that can be made in the future. Currently, our study focuses solely on passenger travel demand, neglecting the potential for drivers to be assigned to restricted areas so as to subsequently cancel orders, leading to a resource waste. Additionally, our spatial analysis does not account for the geographical locations and semantic significance of Points of Interest (POIs). In future research, we plan to incorporate the influence of passenger destinations and consider spatial similarities and POI semantic information as crucial factors to further enhance forecasting accuracy.

Digital transformation is an important way for small and medium-sized manufacturing enterprises (SMME) to enhance their competitiveness. Although the pace of digital transformation of SMME in China has accelerated in recent years, there are still bottlenecks in transformation such as lack of motivation and cognition. In order to improve transaction efficiency and reduce operating costs and risks, enterprises at the nodes of the industry chain collaborate in research and development, production and sales, and influence each other, thus forming an industry chain linkage effect. The development of digital economy has strengthened the linkage effect of industrial chain, and also derived the digital linkage effect. Under the digital linkage effect, digital transformation is not only the strategic decision of a single enterprise, but also affects other enterprises in the industrial chain. Therefore, it is significant to explore the influence of upstream and downstream enterprises (UDE) digitization level on the digital decision-making of SMME based on the perspective of industry chain linkage effect. The theoretical significance in this paper is that based on the linkage effect of the industrial chain, it analyzes the influence of digital synergy of UDE on the digital decision-making diffusion of SMME, and expands the research perspectives of enterprise digital transformation. The practical significance is that it breaks through the limitations of previous literature focusing on the study of digital transformation of SMME from a single perspective, and combines the vertical perspective of upstream and downstream linkage of the industrial chain with the horizontal perspective of competition among peer enterprises, and the results of the study provide new ideas and theoretical support for the government to optimize the digital transformation policy of SMME.
The combination of complex network and evolutionary game can effectively describe the strategy diffusion and the evolution of interactive learning behavior among agents in the network structure, and explain its influencing factors and mechanisms. Therefore, based on complex network theory and evolutionary game theory, this paper constructs an evolutionary game model of complex network, discusses the influence of different factors on the digital decision-making of SMME, and analyzes the dynamic diffusion process of digital decision-making through numerical simulation.
The results show that: (1)The digitalization level of UDE in the industrial chain has a significant impact on the digital decision diffusion of SMME. A higher level of UDE digitalization can force SMME to make a digital transformation. On the contrary, due to the lack of digital cooperation between UDE, SMME that have been digitally transformed can not give full play to their digital advantages to improve their economic efficiency, which indirectly leads to the failure of digital decision diffusion. (2)The linkage effect of industrial chain affects the incentive effect of the government subsidies for the digital transformation of SMME. Only when the UDE in the industrial chain are digitized to a certain extent can the government subsidies generate incentives to promote the diffusion of digital decision-making of SMME. On the contrary, the government subsidy policy will be invalid, and the lower digital cost will become a key factor driving the digital transformation of SMME. (3)The magnitude of reducing cost and improving efficiency has a greater impact on the digital decision-making of SMME than the increased competitiveness and synergy benefits of digital transformation, and all three are exogenous drivers of the proliferation of digital decision-making in the context of higher levels of UDE digitization. (4)The cost of information asymmetry caused by the difference of digitalization levels is an important factor affecting the diffusion of digitalization decision. When the digitalization level of UDE in the industrial chain is higher than that of SMME, if the cost of information asymmetry caused by the difference of digitalization level is low, SMME after digital transformation will play a major role in the diffusion of digital decision-making. On the contrary, the traditional SMME will promote the spread of digital decision-making. Finally, according to the research results and the actual situation, this paper puts forward some policy suggestions to promote the digital transformation of SMME.

Promoting primary healthcare organizations to provide high-quality services, so as to alleviate the difficulty with an appointment with doctors and reduce a high cost of medical treatment, is the key to advancing healthcare reform. To this end, the Chinese government has been constantly promoting the construction of medical alliances, trying to extend high-quality medical resources to hospitals and primary healthcare institutions that are in need of them, so as to solve the contradiction between supply and demand and the problem with a low level of governance of primary medical care. However, the long-established distribution pattern of medical resources is difficult to adjust in the short term, and the imperfect interest coordination mechanism has resulted in a slow progress in the development of medical alliances to strengthen primary healthcare. Hence, the issue of transforming the enduring limited-service capabilities of primary healthcare, encouraging patients to seek care at community level, and establishing a structured and effective diagnosis and treatment service system has attracted significant interest from policymakers and academics.
This study constructs a Hotelling model based on considering patient choice behavior. It delves into the decision-making dynamics of medical alliance participants in their efforts to enhance the quality of primary healthcare services, while also investigating the mechanisms governing patient flows and the overall enhancement of healthcare market performance. Specifically, we establish a medical alliance consisting of a tertiary hospital and a primary healthcare institution, both of which are capable of treating some common and prevalent diseases. However, there exist differences in healthcare service quality, prices and convenience. Patients choose their hospitals based on their preferences. Within the medical alliance, collaboration between the tertiary and primary hospitals is established to enhance the quality of primary healthcare services. Simultaneously, the tertiary hospital aims to decrease the number of patients with common and prevalent diseases to alleviate the patient load. On the other hand, the primary healthcare institution aims to attract more patients to reduce idle costs of equipment and personnel and improve the utilization of healthcare resources. Furthermore, we conduct a comprehensive analysis and comparison to assess how various government subsidy programs impact the decision-making equilibrium of medical alliance members and the overall equilibrium performance of the healthcare market.
The main results are as follows: 1)The establishment of medical alliance can enhance the service quality of primary healthcare facilities and encourage increased patient utilization of the primary hospital. Nevertheless, the extent of effort exerted by medical alliance and the volume of patient return are positively correlated with the initial quality of the primary healthcare institution. 2)Government subsidies promote quality improvement efforts in medical alliance by affecting its performance-cost ratio, thereby reducing its operating costs. However, the extent of cost reduction varies with the differences in the performance-cost ratios of each hospital. When the overload costs of tertiary hospital is disproportionately low, there will be a risk of diverting patients away from primary healthcare institution. 3)When the government subsidizes any member of medical alliance, with the subsidy proportion being the same and patients’ preference for healthcare quality outweighing the unit distance cost, there will be a minimum threshold for the subsidy proportion. Only when it exceeds the threshold does the overall social health surplus improve compared to the absence of government subsidies. Therefore, the government should consider quantifiable operations assessment indicators for medical alliance members, factored in hospital performance and patient preferences. This will better coordinate the impact of financial subsidies on the medical alliance’s development.

In the post-pandemic era, innovation should be more sustainable: taking more social responsibility, taking into account the environment and resources, and paying more attention to low-income groups and promoting social equity and justice. Frugal innovation originates in emerging markets and is oriented to the BoP (Bottom of Pyramid) group in society. The sustainability, economy and inclusiveness of its process are a powerful response to how to innovate under the macro environment of resource scarcity, environmental degradation, the widening gap between the rich and poor, and the shrinking market. Frugal innovation is conducive to the sustainable development of society by adjusting product design, reconstructing the value chain to reduce costs, reducing a resource waste, opening up low-end markets and alleviating poverty.
After reviewing the literature, we find that the existing studies on policy combination and innovation mostly focus on the innovation performance and innovation behavior of enterprises, and the research methods involve ones such as qualitative analysis, empirical research and game models, and pay more attention to the innovation modes of enterprise independent innovation, green innovation, and collaborative innovation, etc., than the relationship between policy combination and enterprises’ frugal innovation. In the current context of resource scarcity, environmental degradation, the increasing gap between the rich and poor, and the shrinking market, how to choose policies and their combination implementation to promote enterprise’s frugal innovation needs to be further explored.
This paper adopts the evolutionary game theory to construct an evolutionary game model of the government and enterprise to investigate the mechanisms of influence of three policies and their combinations, namely, subsidies, penalties, and information inducement, on enterprise’s frugal innovation. We assume that in the state of nature, the process of frugal innovation exists in a system consisting of two parties, the enterprise and government, both of which follow the basic assumptions of limited rationality and imperfect information. In this system, the enterprise has two strategic choices: engaging in frugal innovation and not engaging in frugal innovation; the government also has two strategic choices: taking policy measures and not taking policy measures. By setting the parameters reflecting the cost-benefit of the government and enterprise, the payoff matrix and replication dynamic equation of the evolutionary game between two parties are obtained. We analyze the equilibrium points and stability of the evolutionary game. Numerical simulations are conducted using MATLAB, and the parameters are assigned to describe the laws of the influence of different policies and their combinations on enterprise’s frugal innovation: one policy itself, a combination of two policies, and of three policies.
The following conclusions are drawn: (1)Policy is a necessary tool for government to promote enterprises to engage in frugal innovation. (2)The effects of the three policies and their combinations on enterprise’s strategy choices vary. ①One single policy is not the best choice for government. ②Compared with one single policy, a combination of two policies is more effective, and this combination should be flexible according to the specific situation. When the combination of subsidies and penalties is taken, subsidies should be the main strategy, followed by penalties; when the combination of subsidies and information inducement is taken, the effect of both will be similar, so government can flexibly adjust the combination according to the actual situation in the implementation process; when the combination of information inducement and penalties is taken, the information inducement policy should be the main strategy to encourage, promote and guide enterprises, while supplementing with penalty policies to regulate enterprise behavior. ③In comparison, a reasonable combination of the three policy tools has the best incentive effect on enterprises. The combination of subsidies, penalties and information inducement policy in a proper proportion will more effectively promote the adoption of frugal innovation.