ORIGINAL ARTICLE Optimal location of workstations in tandem automated-guided vehicle systems Amir Salehipour & Mohammad Mehdi Sepehri Received: 18 November 2011 /Accepted: 31 January 2014 # Springer-Verlag London 2014 Abstract The way workstations are located in a tandem automated-guided vehicle (AGV) systems affect the total lateness of the system. So far, almost all studies have focused on either minimizing the total flow or minimizing the total AGV transitions in each zone. This study presented a novel approach to locate the workstations in a tandem AGV zones by developing a new mixed-integer programming (MIP) for- mulation. The objective is to minimize total waiting time of all workstations which is equivalent to minimizing the total late- ness of each zone. Lateness is defined as the total idle time of a workstation waiting to be supplied by an AGV. The proposed MIP formulation is very competitive and has the capability to solve instances of up to 25 workstations to optimality in a reasonable amount of time. Keywords Zone workstation layout . Tandem AGV . Waiting time . Total cumulative flow 1 Introduction Designing efficient handling systems is one of the most im- portant issues in facility design. Tompkins et al. showed that between 20 and 50 % of the overall operational costs was due to material handling [1]. An improvement in the handling system is automated-guided vehicles (AGVs), which are considered as advances to the traditional material handling systems. An AGV is a driverless vehicle which transports materials within a manufacturing area partitioned into cells [1]. The AGV has the advantages of routing flexibility, space utilization, safety, and reduction in overall operating cost [2]. These advantages have led to implementation of AGV in manufacturing and warehousing environments, especially where complex parts are involved with multiprocessing oper- ations. The tandem AGV concept was introduced by Bozer and Srinivasan [3–5]. They defined this system on a grid layout where each workstation is presented as a single point and may represent a machine, or a group of machines, such as a cell or a department. In the tandem AGV system, the workstations are partitioned in such a way that each station is assigned to only one zone where an AGV operates (Fig. 1). In Fig. 1, the circles represent the workstations, and the solid lines represent the guided path of an AGV. The boxes next to each workstation show the input/output to/from each work- station. We refer interested readers to [3–5] for more details on the tandem AGV system. We can break the problem of designing a tandem AGV system into following five problems: & Decision on the number of required zones, and hence, the number of AGVs (problem 1). & Dividing up workstations into zones, i.e., workstation partitioning (problem 2). & Determining workstation layout in each zone, i.e., the optimal location of workstations (problem 3). & Optimal location of transition points between each two zones (problem 4). & Design of AGVs routes’ direction, i.e., clockwise or coun- terclockwise (problem 5). This study focuses on problem 3. Its major contribution is to solve this problem to optimality and to minimize the total A. Salehipour School of Mathematics and Physical Sciences, The University of Newcastle, University Drive, Callaghan, NSW 2308, Australia e-mail: amir.salehipour@gmail.com M. M. Sepehri (*) Department of Industrial Engineering, Tarbiat Modares University, Tehran, Iran e-mail: mehdi.sepehri@gmail.com Int J Adv Manuf Technol DOI 10.1007/s00170-014-5678-x