کنترل موجودی مطلوب از ظروف خالی به اجرا درآمده در سیستم حمل و نقل داخلی

In this paper, we deal with an inventory control problem of empty containers in an inland transportation system. In inland container transportation, freights (containers) are transported between terminal and the customer’s location by trucks, trains and barges. Empty containers are an important logistic resource and shipping companies try to operate and manage empty containers efficiently. Because of the trade imbalance between hub ports, empty containers should be periodically repositioned from surplus areas to shortage areas. However, it is not easy to exactly forecast the demand of empty containers, and we therefore need to build an efficient way to reposition the empty containers. In this paper, we consider a shortage area and propose an efficient inventory policy to control empty containers. We assume that demands per unit time are independent and identically distributed random variables. To satisfy the demand of empty containers, we reposition empty containers from other hubs based on the (s, S) inventory policy, and also consider the lease of empty containers with zero lead time. For the leased containers, we should return the number of empty containers leased to the leaser after the specified period. For a given policy, simulation is used to estimate the expected cost rate and we use the optimization tool, OptQuest® in Arena to obtain the near optimal (s, S) policy in numerical examples.

The transportation demand of containers is rapidly increasing nowadays and the demand for empty containers is also increasing accordingly. Because of the trade imbalance, empty containers should be repositioned between shortage and surplus areas periodically and shipping companies need to have an inventory control policy to reposition the empty containers. Shipping companies reposition empty containers between hub areas, ports and depots. Because it usually takes a long time to reposition empty containers between hub areas and an efficient management of the empty containers is an important factor that can contribute to raising the productivity of shipping companies. Crainic et al. (1993) dealt with the allocation problem of empty containers according to the dynamic and uncertainty of demand. Cheung and Chen (1998) considered how the dynamic container allocation problem can be formulated as a two-stage stochastic network model. They also studied optimization problems for repositioning empty containers and determined how many leased containers are needed at ports. Shen and Khoong (1995) proposed a network optimization model between ports and solved the problem using A Mathematical Programming Language (AMPL). Lam et al. (2007) proposed dynamic and stochastic models for a simple two-port and two-voyage problem. Li et al. (2004), (2007) proposed a new (u, d) policy for the distribution problem of empty containers between ports. In this paper, we consider a port area that needs more empty containers, known as a shortage area (for example, Busan, Korea). Suppose that we should prepare a suitable number of empty containers to satisfy the customer’s seasonally fluctuating demand. To satisfy the required number of empty containers, we can either reposition empty containers from the surplus area with a long lead time or lease empty containers. Thus, we consider the ordering (repositioning) and leasing policy under probabilistic demand and supply, with high and low demand seasons for the demand of empty containers. Holding, leasing and ordering costs are considered and we obtain optimum inventory policies to minimize the expected cost rate (long-run average cost per unit time) by an ARENA simulation.

In this paper, we considered an inventory control problem of empty containers. For probabilistic demand of empty containers with low and high seasons in a hub area, we used the (s, S) inventory policy to order empty containers from other hub areas. While there is a lead time for repositioning, we can lease empty containers immediately. Holding, leasing and ordering costs are considered and the expected cost rate (long-run average cost per unit time) is an optimization criterion. Simulation is used to obtain the expected cost rate and some numerical examples are studied. For given values of model parameters, OptQuest® is used to find the near optimal inventory policy based on simulation results. For further studies, we will consider coordinate inventory problems of empty containers in multi-depot cases and in this paper, an independent and identical distribution is assumed for probabilistic demand, but more complicate stochastic processes can also be used and similar inventory problems can be studied.