Article Preview
TopIntroduction
In the apparel market, there are numerous types of products that have a very short life-cycle and this produces fluctuating demand (Mazziotti & Horne Jr, 1997; Mok, Cheung, Wong, Leung, & Fan, 2013). On the other hand, due to the manufacturing constraints, the lead time in the apparel industry is usually longer in comparison to other industries (Guo, Wong, Leung, Fan, & Chan, 2006; Şen, 2008; Taplin, 1997). Manufacturing in this type of environment requires systems that can quickly adapt to the dramatic changes within the market (Chuter, 1995; Şen, 2008). Moreover, within the US alone, there are approximately 30,000 companies that are actively manufacturing apparel (Statistics of US Businesses, 2013). Given this level of competition, it is crucial that organizations within this sector have efficient manufacturing planning systems.
Similar to other manufacturing systems, scheduling plays a vital role in apparel manufacturing. In fact, in order for an apparel manufacturing system to be agile and responsive to changes in the market, it is necessary to have flexible, reliable, and fast scheduling systems (Bruce, Daly, & Towers, 2004; Tomastik, Luh, & Liu, 1996). Thus, a system developed for this type of environment should consider both scheduling and rescheduling aspects of product planning. This kind of dual-focused system is even more critical within the apparel industry where schedules frequently change due to the volatility of deadlines and demand.
An efficient scheduling system for apparel production should consider two sets of constraints. These constraints can be summarized as either general or apparel-specific scheduling constraints. General scheduling constraints are those that are common across various industries such as starting time, precedency, and set up dependency (Pinedo, 2005, 2015). Apparel specific constraints are mostly related to machine availability. The primary concept behind these constraints deals with pneumatic and steam generator systems. For example, sewing machines are usually connected to pneumatic systems that have a certain air generation capacity. Depending on the capacity, sometimes these systems might not be able to generate enough air to operate all of the sewing machines at once during high-load production periods. A similar constraint applies to the irons, which are connected to a central boiler that generates steam. If more than a certain number of irons work simultaneously, there is often not enough steam pressure for all the irons to function properly.
In this article, a scheduling and rescheduling decision support system (SRDSS) for apparel manufacturing is introduced. While considering general and apparel specific constraints, the proposed SRDSS minimizes the tardiness using a variable neighborhood search (VNS) coupled with Monte Carlo simulation.
The contribution of this paper is threefold. First, an SRDSS framework for apparel manufacturing is proposed. Second, a hybrid VNS-Monte Carlo heuristic algorithm for scheduling and rescheduling is introduced that considers different release times, sequence-based setup times, blocking, and other resource constraints. Third, the SRDSS is implemented in a real-world manufacturing plant and the associated results are analyzed and presented.
The remainder of this article is organized as follows. In section 2, the related literature is discussed. In section 3, the SRDSS architecture is introduced. In section 4, the scheduling/rescheduling method (a hybrid VNS-Monte Carlo heuristic) is explained. In section 5, a case study example is solved, and the corresponding results are presented. Finally, in section 6, the overall conclusion of this research is presented along with a discussion about future research considerations.