In this paper, we study the core acquisition and remanufacturing problem in which the remanufactured products are produced from acquired cores with uncertain quality condition, and are used to satisfy customer demand. Decision-making models are developed to examine the potential value of core sorting and quality grading in the remanufacturing system: a single-period model with deterministic demand, and a single-period model with stochastic demand (i.e., a newsvendor-type model). In each model, both the sorting strategy and the non-sorting strategy are discussed and compared. Our theoretical and numerical results show that: (1) In the deterministic demand case, core sorting is cost-effective only when the unit sorting cost is below a threshold value and the unit acquisition cost falls into a specific interval. Furthermore, in the case with two quality grades the adoption of sorting strategy with respect to the expected fraction of high-quality cores may be non-monotone: an initial increase in the expected fraction of high-quality cores may motivate a switch to core sorting, however, further increase in the expected fraction may motivate a reverse switch; (2) Similarly, in the stochastic demand case, the sorting strategy also becomes unattractive when the unit sorting cost is sufficiently high. In addition, the value of core sorting will be better off under more fluctuating demand for remanufactured products if the sorting strategy is the dominant strategy. Otherwise, it will be worse off.
1 October 2018 in Tehnicki vjesnik - Technical Gazette
Travel demand forecasting in terms of origin-destination (O-D) matrix estimation on transportation networks is an important topic in transportation research. The motivation of each trip is the user’s desire of conducting a compulsory...
The flow pattern on a given transportation network at a given moment results from many users’ travel decisions which are made for some purposes, for example, participating in necessary activities such as work, eating and shopping. Consequently, the explicit modelling of the interaction between users’ activity and travel choice behaviour serves as a basic building block for long-term transportation planning and management. In this paper, an activity-based network user equilibrium model is proposed to study the dynamic activity-travel scheduling problem with multiple classes of users under time constraints. A simple supernetwork representation approach is introduced to generate the activity-journey-network (AJN) which expands the basic transportation network in both time and space dimensions. With the supernetwork representation, the dynamic activity-travel scheduling problem is transformed into a static network flow assignment problem. A heuristic algorithm is developed to find the path with the maximum utility in the AJN from the start node to the end node for each user. A numerical study is conducted to illustrate the application of the proposed model and solution algorithm for several transportation networks including large-scale real-world networks. It is shown that both the individual’s travel choice and group travel behaviour in a transportation network can be well studied by the proposed model.