Optimization of operation and changeover time for production planning and scheduling in a flexible manufacturing system q Kanchan Das a , M.F. Baki b, * , Xiangyong Li b a Technology Systems Department, East Carolina University, Greensville, NC 75858, USA b Odette School of Business, University of Windsor, 401 Sunset Avenue, Windsor, Ont., Canada N9B 3P4 article info Article history: Received 29 June 2006 Received in revised form 6 May 2008 Accepted 3 June 2008 Available online 11 June 2008 Keywords: Process planning Sequencing Flexible manufacturing system Machining Integer programming Part grouping Machine loading Tool allocation abstract This paper deals with the production planning problem of a flexible manufacturing system. It specifically addresses issues of machine loading, tool allocation, and part type grouping with the intent of developing an operation sequencing technique capable of optimizing operation time, non-productive tool change times, and orientation change times when processing a group’s design features. A hierarchical approach has been adopted to determine the part groups – depending on the operation, tool change and orientation change times at the upper level. At the next level, we sequence the operations of the part groups. Integer programming models are formulated to group the parts and to address the operation-sequencing prob- lem. The model is illustrated with an example related to an auto engine cylinder head machining plant. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Machines in modern, flexible, manufacturing systems are capa- ble of performing all planned operations. The inherent efficiency of a flexible manufacturing system (FMS), combined with additional capabilities, can be harnassed by developing a suitable production plan. As Stecke (1983) mentions, in order to best utilize an FMS’s capabilities, a careful system set up is required prior to production. This paper considers the specific production planning problem of complex parts, similar to an automobile engine cylinder head, which combines design features (DFUs) equivalent to part types, and unit manufacturing features (UMFs) equivalent to operations. The DFUs are located at different locations and, in some situations, at different faces. Although the tool change time is minimal in modern FMSs, the movement of tool heads from one DFU to the next, combined with the time needed for tool retracting and posi- tioning, makes each tool change considerably time consuming. In addition to the tool changes, orientation changes required to reach the DFUs at different faces of the cylinder heads make the system more complex. We follow the general framework introduced by Stecke (1983) as a guideline, but resort to a system-dependent planning approach more suitable to our problem. Among the FMS planning models, the machine loading one is most frequently cited. Part grouping is the next most studied mod- el, as in Hwang and Shogun (1989), Kulkarni and Kiang (1995), Liang and Dutta (1993a, 1993b), Rajagopalan (1986), Sawik (1990), and Stecke and Kim (1988, 1991). Mohammed, Kumar, and Motwani (1999), Mukhopadhyay, Maiti, and Garg (1991), Rajagopalan (1986), Sodhi, Askin, and Sen (1994) consider an approach that combines part grouping and machine loading in conjunction with tool loading. Part grouping, machine loading, and tool provisioning for part groups are, in essence, linked. This paper takes a joint approach, collectively addressing these three problem areas for a more comprehensive solution. To implement the above-mentioned joint approach, this paper introduces a methodology for considering operations, tool change and orientation change times that addresses the FMS planning problems of part grouping, machine loading, and tool allocation. Assigning machines to a part group equipped with the required number of tools is a crucial part of both the machine loading and tool provisioning phases. As discussed, modern FMS machines are capable of performing all operations. However, a different set of tools gives a machine a different capability as required by the planning model to process a planned part, mix, or group. The out- put is then taken from this planning model, and a detailed opera- tion schedule is generated in the later stage. Section 2 reviews the relevant literature. Section 3 presents the mathematical programming models for the part grouping, machine 0360-8352/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.cie.2008.06.001 q This manuscript was processed by Area Editor Gursel A. Suer. * Corresponding author. Tel.: +1 519 253 3000; fax: +1 519 973 7073. E-mail address: fbaki@uwindsor.ca (M.F. Baki). Computers & Industrial Engineering 56 (2009) 283–293 Contents lists available at ScienceDirect Computers & Industrial Engineering journal homepage: www.elsevier.com/locate/caie