It is more difficult for the investor to monitor entrepreneur's moral hazard under growth risk due to the lack of information. To solve this problem, present study employed robust method to construct an equity financing model of cautiously optimistic entrepreneur facing growth risk, and to investigate the effect of agency problem on equity financing. Results indicate that the enterprise becomes more conservative under agency problem, which inhibits the equity financing and rapid development of the enterprise. Increasing the mean growth, reducing the wholesale price, or improving the enterprise valuation under the agency problem can improve the risk tolerance of the enterprise. In addition, handing over the management rights to the entrepreneur does harm to investor's benefit when investing the enterprise with high valuation, high growth mean or low growth risk.
This paper studies the two retailers horizontally differentiated products assortment problem under market competition. We adopt multinominal logit (MNL) model to characterize the consumers' choice behavior and establish the basic model to obtain the optimal assortment decision under the market independent circumstance. Then we classify the competitive circumstance into four kinds of scenarios. In each of these scenarios, we study whether retailers enter the market sequentially affects the final results, and compare the results of independent and competitive circumstances. The conclusion shows that the degree of product variety in the competitive circumstance is no less than that of the independent circumstance. Some products which have two retailers' profit space is a necessary condition for retailers to choose common products. Some products which are not in the optimal assortment decision under the market independent circumstance could be profitable is a necessary condition for the competitive circumstance has more product variety. When retailers enter the market sequentially, information asymmetry does not affect the result.
To sell the deteriorating products timely, distributors want to use different marketing strategies to maximize their profits, i.e. providing the credit period to increase the enthusiasm of buying or taking the delayed delivery strategy to reduce the deteriorating cost. However, for distributors, how to select the optimal strategy is a genuine problem. This paper constructs an inventory model based on a delay-in-payment strategy for deteriorating products and an inventory model based on the delayed delivery strategy, respectively. For such two models, optimal strategies are proposed, respectively. Furthermore, by comparing the two optimal strategies, we find that when the ratio of the sale price and the wholesale price is relatively low while other parameters are given, distributors can take delay-in-payment strategy, otherwise, retailers can take delayed delivery strategy. Finally, by utilizing the numerical examples, we discuss the disturbance of related parameters for the optimal strategy, and also put forward several reasonable suggestions for the distributors.
With further study of game theory, evolutionary game model has been widely used in the analysis of many social phenomena and economic problems. In evolutionary game model, strategy updating rule is introduced into the state transition equation, and the corresponding Markov chain is obtained to study the evolutionary state of population. When the Markov chain has no absorbing states, the average abundance function is used to study the evolutionary state of population. The extended average abundance function is derived by analyzing the stationary distribution of Markov chain using aspiration driven rule in the strategy updating rules. At the same time, by applying the multi-player evolutionary game model to snowdrift evolutionary game, extended average abundance function of multi-player snowdrift evolutionary game model is obtained. By means of numerical analysis, the influence of corresponding parameters on the average abundance function is calculated and analyzed, and how the variation of parameters affects the behavior of enterprises in the game is studied with a specific case. The research shows that the proportion of cooperators will be increased by changing relevant parameters, and this conclusion points out the direction for how to regulate the corresponding parameters to promote cooperation in practical application.
This paper constructs a theoretical model to analyze the mechanism of factor resource misallocation on enterprise innovation, and empirically tests the effect of factor resource misallocation on enterprise innovation by using the data of manufacturing companies from 2012-2016 in China. It is found that there is a cumulative effect of enterprise innovation, and factor resource misallocation is an important factor affecting enterprise innovation. However, labor and capital resource misallocation have different effects on enterprise innovation. On the whole, labor resource misallocation significantly promotes enterprise innovation. Regionally, the effect of labor resource misallocation on enterprise innovation has decreasing tendency in central, east and west region. Capital resource misallocation significantly inhibits the enterprise innovation in the central region. Based on the nature of equity, labor resource misallocation promotes enterprises innovation in state-owned enterprises and private enterprises, and capital resources misallocation inhibits enterprise innovation in private enterprises and foreign enterprises. Based on the view of industry, different factor resource misallocation on enterprise innovation exist industry difference. In addition, enterprise scale, capital density, enterprise age and market concentration also affect enterprise innovation in varying degrees.
In order to construct a more reasonable and optimal contract model, this paper assumes that the principal is time-inconsistent preferences and sophisticated in the classical continuous-time agency model, where the hyperbolic-discounting functions characterize time-inconsistent preferences. This paper studies the effect of the principal's time inconsistency on her value function and optimal payoff boundary. Compared with the classical continuous-time agency model, the extended model shows that the sophisticated principal's value function and optimal payoff boundary decrease as the degree of time-inconsistency exaggerates. Namely, to remedy the uncertain risk of the time-inconsistent preference, the time-inconsistent principal tends to pay out lower compensations earlier and more frequently than a time-consistent peer, so that the agent chooses to truthfully report cash flows truthfully earlier. It indicates that the principal's time-inconsistent preferences have a great impact on the optimal contract.
As terrorists have a strong preference in attacking civilians in densely populated areas, terrorist attacks can easily cause serious consequences. In order to improve the rescue efficiency and reduce the loss of attacks, the State can locate emergency facilities in the transportation network. Since the one with information advantage is usually dominant in the game, the State can also mislead terrorist's actions and improve her utility by hiding some location information. First, we describe the research problem and address it to a bi-level programming model in which terrorists' bounded rational behaviors are depicted according to the random selection theory. Second, both exact solution method and genetic algorithm are proposed, and they are applied in a real-world case study of Kashi area. The result shows that: when the State is able to calculate the rationality of terrorists, compared with disclosing all location information, hiding some information is more beneficial for reducing the expected loss, and the degree of information hiding is highly related to the rationality of terrorists. Conversely, when the State isn't able to calculate the rationality of terrorists, the information hiding strategy always plays a more effective role in the case that the terrorist's rationality is underestimated.
Emergency facilities are the carrier of emergency rescue, the scientific and reasonable location is related to the urgency of emergency rescue and the timeliness of distribution of emergency resources, it is of great strategic significance to study the location of emergency facilities and the decision-making of emergency resource allocation with obstacle constraints. From the perspective of demand area and the emergency service quality of emergency facilities, a location-allocation optimization model of emergency facilities based on obstacle constraints, capacity and safety stock constraints was constructed. The paper introduced a safe stock mechanism and considered the multiple constraints such as time, economy and geographic blocking, the decision-making process of the location and distribution was analyzed to establish a location-allocation scheme for emergency facilities. Coupled grey wolf optimizer (GWO) and visual convex point method was presented to solve the model, numerical examples show that the proposed algorithm can efficiently optimize path around barriers, and get the optimal location allocation scheme under different time satisfaction preference of various regions. The research results will provide a model and methodological design for the location and resource allocation of emergency facilities.
Congestion pricing has been considered as one of effective means for mitigating urban traffic congestion. In this paper, we propose a link-based congestion pricing scheme which utilizes the origin-destination information of vehicles in traffic networks. We have roundly compared this scheme—the minimum total revenue pricing based on link-differentiated (MTRP-LD)—with other three ones fully reported in literature, namely the marginal social cost pricing (MSCP), the minimum total revenue pricing based on link (MTRP-L) and the minimum total revenue pricing based on path (MTRP-P). It is proved that all these four schemes can realize the flow pattern of system optimum, and that the total revenue is only related to the traffic demand and the minimum path cost when tolled user equilibrium is achieved. MSCP generates the highest toll revenue, then MTRP-L, MTRP-LD and MTRP-P in order. MSCP and MTRP-L can be implemented in an anonymous way, while MTRP-P requires the complete path information of vehicles. MTRP-LD proposed in this paper which can reduce the total toll revenue greatly like MTRP-P, is link-based for implementation but needs all vehicles' origin-destination information. Numerical results obtained from a multiple origin-destination network verify our analytical results.
Based on the 176 airports data of China from 2005 to 2014, an analytical method called the standard deviational ellipse method in GIS is firstly introduced for spatio-temporal evolution analysis of air passenger and cargo system. Then, the spatial agglomeration and disparity of passenger and cargo is identified by space random test and probabilistic distribution. Finally, we use spatial similarity method to identify the driving factors of air transport. Results show that the principle orientation of airport passenger is westward, which is similar with that of airport cargo volume, besides, the moving of cargo volume spatial pattern is faster than passenger volume spatial pattern. The spatial disparity of passenger and cargo volume is gradually decreased, presenting the development tendency of balance. The geographical distribution of air transport agglomeration is obvious, and GDP per capita is the main driving factor of air passenger, whereas regional GDP is the main driving factor of air cargo. Results of this study are useful for hub-airport siting and airport layout decision in China.
With the Chinese population ageing, home-health-care industrials become more and more important. Characteristics of HHC in China differs from other countries, such as the intensive distribution of senior people, whose residence centralizes in community. This thesis's problem was based on the field of home health care in huge communities in China, studying how to decline caregivers' moving distance and optimize the daily worksheets. According to three combinations of nurses' skill levels and customers' demands, service time of elders were assumed to obey normal distributions with different mean values and variances so as to analyze the impact from fluctuation of demands on algorithm and decision-making. In real business elders generally make appointments in advance, and appointments are descripted mathematically by time-windows. The impact of different types of appointments and appointments with different intervals were analyzed. The authors optimize the function of transition-probability-calculating in original ant-colony algorithm, using multiple instances to test the algorithm. The result shows that with replacing manual work, computer and algorithm shapely cut down the time cost of decision-making.
To remedy the deficiency of the single algorithm in optimality and diversity when solving the multi-objective flexible job shop scheduling problem, we proposed a multi-strategy integration Pareto artificial bee colony algorithm (MSIPABC). First, this algorithm employs the hybrid heuristic strategy to initialize the food source colony to obtain the initiation colony with higher quality. Then the employed foragers use multiple local search operations to explore new food sources around the current food source. The onlookers use the tournament rules to select the superior food sources and perform crossover operation and the critical-path-based neighborhood search to further enhance the optimization search of the algorithm. At last, the scouters reconstruct the repetitive solutions, ensuring the diversity of the food source colony. The algorithm adapts multiple search strategies, realizes the autonomous and cooperative search of artificial swarm and hits a balance between global search and local search. The proposed algorithm is proved to be superior both in solution quality and diversity.
In order to solve the vehicle scheduling problem with traffic disruption delay, a two-stage disruption management method based on multi-phase quantum particle swarm optimization (MQPSO) is proposed. Firstly, a mathematical model of the problem is established to minimize the time window deviation and minimize the distribution cost. Next, the possible disruption management mode of distribution vehicles is summarized and classified and the route scheduling method is proposed on the basis of selected mode. At last, the simulation experiments are executed based on the Solomon example to test the performance of the proposed method. The effectiveness and practicability of the proposed method is verified through the comparison and analysis with rescheduling method.
From the cognitive perspective, this research theoretically analyzes the formation and impact of conflicts in the process of strategic change, proposes the hypothetical model of strategic change cognition, strategic change response as well as conflict relationship, and also carries out an empirical study. Based on our results, it shows that strategic change cognition and strategic change response are positively correlated, and conflict plays an intermediary role between them.
The application of grounded theory to explore the doctor-patient trust against the deep-seated factors, and found that the five main categories of patients' goodwill, patient's integrity, patient's ability, medical inherent characteristics and institutional social context have significant influence on the doctor-patient trust. The patient medical image is the doctor-patient trust against the pre internal variables, the inherent characteristics of internal medicine situational variables of doctor-patient trust violation, the system of social factors is the external situational variables of doctor-patient trust violation, they affect the role and path of doctor-patient trust violation is not consistent. On this basis, we explore the component factors of the 5 main categories and their mechanism of action against the breach of trust between doctors and patients. The study can provide effective policy for the government to repair the trust between doctors and patients, and provide specific policy ideas and implementation paths.
The classical hierarchical covering location problem (HCLP) is the problem to find locations within a limited budget to provide hierarchical services. In general, the ability of a hierarchical facility to export services to an affected area depends on its range of service radii and is not affected by service availability. Instead, this requirement can come from different facilities with different service availability in the hierarchical network. We designed a hierarchical network of hybrid service availability and constructed an integer-programming model for the hybrid hierarchical backup coverage location problem by discussing the number of hierarchical facilities as quantitative and variable, and developed the meta-heuristic algorithm. It is shown that the optimization model meets the needs better in coverage capability, while the backup coverage capability and system cost are not always better than the single-type hierarchical network; and the suggested heuristic yields high quality solution in a reasonable computation time.
To the dynamic evaluation problems with interval uncertainties, we further discuss the stochastic clustered solution from the aspect of stochastic simulation and develop other types of evaluation results. Firstly, the preference ratio matrix, representing the degree of which one alternative is superior to another one, is introduced, and the simulation solution and the associated simplified method for this matrix is developed. Then other types of results, including the possible ranking with preference probability, the dynamic estimated value for alternatives indicating their development trend, are provided based on the preference ratio matrix. At last, an example related to the dynamic evaluation of the automated assembly line for weapon building is introduced to illustrate the solution of these different types of results and give some analysis. This research can not only be more consistent with the dynamic uncertainty environment, but also further enrich the evaluation results, which can provide more information for the understanding of the alternatives' development trend from different aspects.
This paper investigates an approach to record incremental grouping based on transferred similarity for large data sets. The paper first analyzes how to gradually calculate similarity between records, then proposes how to construct reference group based on sorting key, how to incrementally update reference group based on transferred similarity, and how to perform incremental updates in reference group based on union-find, finally proves the feasibility and efficiency of the proposed method through experiments. Experimental results show that the proposed method can improve grouping quality and improve grouping efficiency more than traditional methods. There is no detailed analysis of the data quality problem existing in the attribute value itself, and there is no design of the sorting key generation algorithm. The proposed method can not only help solve the problem of missing record pairs in data cleaning, information integration and management, but also has advantages such as better scalability, reusability, and freedom from the domain, because it only designs algorithms from the perspective of pure data processing.
Aiming at the problem that the cloud computing network is prone to load imbalance, for the cloud computing network node heterogeneity, the differences of the allocation of resources and users demand uncertainty factors, on the basis of analyzing the fuzzy temporal variation characteristics of cloud node load, the dynamic load balancing (DLB) model of cloud computing network based on intuitionistic fuzzy time series (IFTS) is proposed. Meanwhile, the self-balancing algorithm for cloud node computing resource based on IFCM is proposed, the virtual machine scheduling mechanism based on mixing together the active control based on IFTS prediction and the virtual control based on feedback is designed, and the cloud computing network dynamic load balancing strategy is presented, which effectively enhances the intelligent management level of cloud resource pool, improves the overall performance of cloud computing system. Finally, the validity and superiority of the proposed method are verified by a typical example.
The scheduling algorithm of FTSA (feedback mechanism based two-stage switch architecture) must be completed within the crossbar reconfiguration time. Such harsh time limitation makes it impossible to achieve its very excellent theoretical performance. For this problem, a novel switch architecture called AFTSA (adjacent-port scheduling information and feedback mechanism based two-stage switch architecture) was proposed in this paper. The front-feedback mode was introduced into AFTSA, which enabled the input port to obtain the queue information of the destination middle-port one time-slot ahead. Furthermore, a dedicated communication links between the adjacent input ports was proposed and thus makes any input port can distribute the local scheduling results to its adjacent input port. Based on the scheduling results, the middle-queuing information can be corrected. Every time slot, a scheduling algorithm using the adjudication model will be executed to choice a cell to be forwarded to the middle-port in the next time slot. Theoretical analysis shows that the AFTSA can provide close to a time slot of execution time for the scheduling algorithm and thus will enhance its feasibility of practice. Beyond that, AFTSA has the equivalent delay performance as FTSA does in the same switching environment.
Study on the multi-state reliability modeling, analysis and assessment, which plays an important role in controlling time-varying process reasonably and programming supportability activity effectively for the big complex equipment system. In this paper, a universal method adopted to analyze and assess the accumulative characteristic of reliability is given with respect to the multi-state complex equipment systems under time-varying demand condition. Meanwhile, the logic reward matrix system is established and the accumulative logic reward is calculated during work period. Research shows that method adopted in this paper has better robustness, which can meet different constraint condition and assessment request. And that also has better adaptability and routinization, which can be used to assess any multi-state systems with high state dimension uniting with the universal generating function (UGF) technology and can easily be realized by the computer programming considering the matrix vector model. Moreover, the result can be applied to analyze and assess multi-state system reliability in other fields, such as energy industry, information industry and machinery industry, and has important directive to equipment support system construction.
For analyzing the component states which are multi-states and not comparable in the phased-mission system (PMS), the hierarchical manner is presented for the two levels including system and components. The continuous time Markov chain (CTMC) is used to depict the dependence and transition of states, and the multi-valued decision diagram (MVDD) to model the structure function of system. In order to deal with the states which are not comparable in PMS, the states are grouped as three types, i.e. in group, among group and no restriction. In the case of different transitions, the joint probability model of component states in all stages is developed based on CTMC, and the reliability models of MS-PMS with the states which cannot compared are presented. Finally, a practical example about flying mission system is used to illustrate the practicability, and the results indicate that the layered and grouped approach can compute the reliability of MS-PMS effectively.
The research of the sensitivities of the wind power system' reliability indices with respect to the distribution parameters of the wind power forecast error (WPFE) may help the reserve schedule, while the unknown distribution of the WPFE makes it difficult to calculate the sensitivities. The non-standard third-order polynomial normal transformation (NSTPNT) method is novelly proposed. And the analytical expressions of the polynomial coefficients are derived. Based on the NSTPNT, the sensitivities of the reliability indices with respect to the distribution parameters of the WPFE, including the expectation and standard deviation, are established respectively. The numerical results verify the accuracy of the proposed methods. The effects of the reserve capacity and the standard deviation of the WPFE on the reliability sensitivity are analyzed.
The INS/CNS integrated navigation technology, having the characteristics of the whole-process autonomous controllability and non-accumulation time-varrying error, remains a hot and challenging issue in the field of intermediate-range missile navigation. There has been an increased research interest on this issue around the world. Therefore, investigating this issue is of great theoretical and practical importance. On basis of the current research status of the technology, the technical principles and some key techniques such as wild-field starlight sensor, star patterns identification, starlight refraction are analyzed and summarized in this paper. Finally, the development trends of the INS/CNS integrated navigation technology on intermediate-range missile are pointed out.