Co-ordination, warehousing, vehicle routing and deliverymen problems of supply chains

Abstract

Compared with traditional business management aiming to pursue single entity's own maximum benefit or minimum cost, supply chain management, which serves the market from a more systemic perspective, focuses its attention on achieving good responsiveness as well as economy performance through proper coordination of all the participants across the business functions of the supply chain, including procurement, manufacturing and distribution, etc. With the bloom of information evolution, more attentions have been given to increase the degree of coordination among multiple functions within the supply chain. Although, the more decision modules integrated in one co-ordination mathematical model, the better performance of the supply chain would be achieved, it is at the expense of computational time for searching the optimal solution due to the complexity of the model. Hence, our co-ordinated systems would focus on integrating two or more of the following decision levels: procurement, production, inventory, warehousing, vehicle routing and delivery men routing. This thesis proposes and develops the mathematical models and solution methods for four supply chain coordinating systems. (i) Asynchronized cycles single-vendor multi-buyer supply chain model involving clustering of buyers with long and short cycles is proposed. The ordering cycle of each buyer is either an integer multiple or an integer factor of the vendor's production cycle. The buyer-clustering mechanism, in which the ordering cycles adopted by buyers are allowed to be larger than the vendor's production cycle, increases the flexibility of the system, which reduces the total system cost. The effectiveness of this clustering synchronized cycles model is also analyzed. (ii) An integrated production-warehouse location-inventory (PWLI) model is proposed. In this model, decision variables of warehouse location, production schedule and ordering frequency, are integrated in one model and determined simultaneously by minimizing the total system cost. Meanwhile, a synchronization mechanism is implemented to the system so as to coordinate inventory replenishment decisions. Numerical experiments have been carried out to illustrate the performance of this co-ordinated model. (iii) An extension of the synchronized cycles PWLI model is proposed. In this extended model, deliveries are modeled by a set of heterogeneous vehicle routing problems instead of a fixed cost for each order. Numerical experiments have been carried out to illustrate the performance of this co-ordinated model. (iv) An integrated model for multi-depot vehicle routing and delivery men problem is studied. This model incorporates a distribution network of multiple depots, multiple parking sites and multiple customers linked by trips of a fleet of homogeneous vehicles and a number of delivery men assigned to the vehicles. The objective of this model is to determine the number of delivery men assigned to each vehicle and the routing of vehicles and delivery men so as to minimize the total relevant costs involved in the two levels.