An analytical method for configuring fixed-path, closed-loop material handling systems†

SUMMARY This paper presents an analytical method for configuring the network of a fixed-path, closed-loop material handling system. The method is applicable to towlines, automated monorail systems (AMS), automated guided vehicles systems (AGVS), and other equipment types. The purpose is to make good initial decisions with respect to adding shortcuts (cutbacks), adding off-line spurs, and their length, so that of the thousands or millions of possible network configurations only a very small number need subsequently to be examined by simulation. The approach taken is to develop volume-delay curves for load-carriers passing a load/unload station, based on average transport requirements. Using this information we develop, as a function of the number of carriers moving past a station, cost curves that include carrier costs and spur costs, for the two possible situations at each station: build or don't build a spur. A mainline blocking tolerance is used to determine the spur capacity for the build situation. After linearization, the segments of the cost curves are represented by a set of mathematical network flow arcs, with some arcs representing fixed costs and others representing variable costs. We minimize the sum of load-carrier costs and spur costs in this flow network by using a general purpose integer programming algorithm. The problem solution indicates where off-line spurs and shortcuts should be constructed, the number of carriers required, their recommended routeings, and the average flow on each part of the system. The trade-off between the costs of alternative networks, influenced by shortcuts, spurs, load-carrier costs, and by routeing and congestion, is demonstrated by a small example with seven stations. The capability of the method is shown by solution of a different example, one involving 32 build-don't build decisions. Future work in this area is planned with two objectives in mind: simplifying and speeding up the analytical method presented in this report; and linking the analytical method with a simulation procedure.