Jinsheng Shen’s research while affiliated with Beijing Jiaotong University and other places

Study on the intrinsic properties and evolutional mechanism of urban public transit networks (PTNs) has great significance for transit planning and control, particularly considering passengers’ dynamic behaviors. This paper presents an empirical analysis for exploring the complex properties of Beijing’s weighted bus transit network (BTN) based on passenger flow in L-space, and proposes a bi-level evolution model to simulate the development of transit routes from the view of complex network. The model is an iterative process that is driven by passengers’ travel demands and dual-controlled interest mechanism, which is composed of passengers’ spatio-temporal requirements and cost constraint of transit agencies. Also, the flow’s dynamic behaviors, including the evolutions of travel demand, sectional flow attracted by a new link and flow perturbation triggered in nearby routes, are taken into consideration in the evolutional process. We present the numerical experiment to validate the model, where the main parameters are estimated by using distribution functions that are deduced from real-world data. The results obtained have proven that our model can generate a BTN with complex properties, such as the scale-free behavior or small-world phenomenon, which shows an agreement with our empirical results. Our study’s results can be exploited to optimize the real BTN’s structure and improve the network’s robustness.

This paper studies how to coordinate a supply chain consisting of one supplier and one retailer with nonlinear price-demand function under market demand disruptions. Different from conventional research on channel coordination, it mainly focuses on how the demand disruption affects the coordination of the supply chain in which production cost is a convex function of production quantity It shows that the original production plan of the supply chain has certain robustness under centralized-decision system. For the decentralized-decision system, the coordination scheme designed for initially estimated market demand needs to be revised. To resolve this issue, an adjusted revenue sharing contract is presented, and it is proved that the supply chain can be fully coordinated by the adjusted revenue sharing contract.

We demonstrate the problem of how to make the emergency response decision of the supply chain when facing disruptions from the perspective of system coordination. For a two-stage supply chain consisting of one supplier and one retailer who faces stochastic effort-dependent market demand, the quantity-discount contract can help coordinate the supply chain, but this coordination may be broken off by an emergency event, which may make demand disruptions; i.e., it makes the demand distribution different from the assumption needed to achieve the original supply chain's coordination. This paper gives an optimal strategy for a supply chain with effort dependent demand for the disruption, proposes adjusted quantity discount contracts which can coordinate the supply chain under the disruptions, and provides some managerial insights. A numerical example is given to illustrate the theoretical results.

The development of modern logistic industry in the Area along the New Eurasian Land Bridge (NELB) is significant to promote the socio-economic development of the Chinese mid-west region. Based on the analysis of status achievement and existing problems of logistic cooperation&development (C&D) in the Area, this paper discusses its development trend with forecasting the relationship between economy and logistic development. And according to SWOT analysis, it makes a development strategy with applicability in long views, and designs the whole layout and integrated logistic cooperation pattern. The result could be used for reference to develop the modern logistic industry in the Area.

This paper develops an analytical model that explores how important supply chain parameters affect the cost savings to be realized from collaborative initiatives such as vendor-managed inventory (VMI). Results from the model show that the optimal replenishment frequency with VMI is greater than that without VMI, but the optimal shortage quantity and the optimal order quantity for buyer with VMI is smaller than those for buyer without VMI. Results also show that under a certain condition, the total inventory cost with VMI is lower than that without VMI and these benefits are disproportionally distributed between buyers and suppliers.

As an important phase in the plan of Logistics Park, traffic volume forecast and traffic assignment are always the research focus.In this paper, the improved traffic volume non-repeated coefficient method was introduced to forecast the Traffic Volume of Logistics Park, based on the results, system optimization assignment model was used for traffic assignment of Logistics Park with the aid of TRANSCAD. The experimental result may be provided for the road network planning of Logistics Park, making the traffic organization of Logistics Park more effective and optimal. Then, it will bring tremendous economic benefits and social benefits.

In this paper we study the impact of introducing a mobile channel into a supply chain. Thus, the supply chain owns an electronic channel, a mobile channel and a traditional channel. We model the context as a Stackelberg game. Through numerical experiments, we find that as mobile channel becomes more distinct from existing channels and becomes cheaper, consumers and manufacturer will be better off. We also find the manufacturer's inclination to mobile channel will not eliminate the traditional channel. However, the manufacturer's mobile channel has to keep a balance between attracting consumers for retailers and for itself.

Bus route evaluation (BRE) of the public transit system in Beijing is requisite decision to regulate existing bus routes and reallocate transit resources for sustain operation. Since there wasn't a systematic BRE system for Beijing, firstly, this paper develops comprehensive BRE standards according to the status quo of Beijing's transit system, which includes 3 categories, 20 criterions in total. Secondly, based on method of Analytical Hierarchy Process (AHP), a BRE model is built, which integrates quantitative and qualitative attributes of routes. Finally, in terms of passenger flow data investigated, a case study of BRE among No. 7, No. 15, and No. 334 bus routes is carried out. It shows that the evaluation result by this BRE system is consistent with the actual condition of the transit system in Beijing. It is practicability and maneuverability to apply this model to regulate and optimize bus routes in Beijing.

A synchronized optimization model that integrated timetabling, vehicle scheduling and crew scheduling was developed according to the regional bus operating system and the multilevel programming in this paper to deal with the relative ineffectiveness of the traditional single-line and sequential bus operating model from the perspective of both utilizing the company's resources and meeting the passengers' trip demands. Simultaneously, the satisfaction criteria were initialized to select a set of the satisfied solutions instead of an only optimized outcome for the succeeding model to decide. Finally, an experiment with the Multilevel Nesting Taboo Search (MNTS) was given, and the computational results demonstrated that the model was reasonable and that the algorithm was efficient and feasible.

Electric bus scheduling problem can be defined as vehicle scheduling problem with route and fueling time constraints (VSPRFTC). Every vehicle’s travel miles (route time) after charging is limited, thus the vehicle must be recharged after taking several trips and the minimal charging time (fueling time) must be satisfied. A multiple ant colony algorithm (ACA) was presented to solve VSPRFTC based on ACA used to solve traveling salesman problem (TSP), a new metaheuristic approach inspired by the foraging behavior of real colonies of ants. The VSPRFTC considered in this paper minimizes a multiple, hierarchical objective function: the first objective is to minimize the number of tours (or vehicles) and the second is to minimize the total deadhead time. New improvement of ACA as well as detailed operating steps was provided on the basis of former algorithm. Then in order to settle contradiction between accelerating convergence and avoiding prematurity or stagnation, improvement on route construction rule and Pheromone updating rule was adopted. A group feasible trip sets (blocks) had been produced after the process of applying ACA. In dealing with the fueling time constraint a bipartite graphic model and its optimization algorithm are developed for trip set connecting in a hub and spoke network system to minimize the number of vehicle required. The maximum matching of the bipartite graph is obtained by calculating the maximum inflow with the Ford–Fulkerson algorithm. At last, an example was analyzed to demonstrate the correctness of the application of this algorithm. It proved to be more efficient and robust in solving this problem.