Using operational flexibility (lot sizing, scheduling and sequencing) to improve customer responsiveness in a manufacturing system.

Abstract

Sequencing (or resequencing) products in a <italic>blocking flowshop</italic> is a valuable way to make simple, low-cost operational changes to significantly alter system performance without additional capital costs. It often occurs that planned operations in manufacturing systems have to be changed because of changing circumstances. One of the most critical sources of variation stems from customers changing their orders after production has begun; this trend will play an increasingly significant role in the manufacturing industry. Thus, the goal of this research is to assess how product sequencing can enable manufacturers to provide cost-effective delivery of <italic>the right product in the right amount to the right customer at the right time </italic>. In our research, we consider two different approaches to incorporate variability in planning. In the first section of this research, we expand the existing deterministic models to incorporate surrogates of responsiveness. During analysis, it is observed that blocking flowshop scheduling problems often possess large numbers of equivalent optimal solutions. However, within this set, there may be differences in other characteristics. Therefore, metrics are identified to evaluate responsiveness. Models are constructed to consider both makespan and responsiveness. Computational experiments are conducted to analyze the schedules identified when different objectives are considered. In the second stage of this research, a simulation model has been developed to more accurately analyze the impact of variability. Our focus is on developing a methodology for sequencing products to achieve a tradeoff between operational costs and costs associated with customer responsiveness. Our methodology aims to determine cost-effective and responsive sequencing approaches for customer order changes as well as machine breakdowns in varying industries. At the core of this methodology is the notion of a <italic>policy</italic>, which is a set of rules for controlling product sequencing during production. A responsiveness framework is developed to help users assess and analyze a variety of policies. Computational experiments are conducted under high volume, low mix and low volume, high mix environments to generate insights. This research shows that product resequencing and more complex policies are effective methods especially in high volume, low mix environments because of significant risk associated with over-or-under producing.