Development of production and inventory control policies in imperfect production systems

Abstract

The production and inventory control problems in manufacturing systems have attracted a large number of researchers in the past few decades. Holding inventory, such as raw material,tools and assemblies, not only ties up capital but also generates an associated carrying cost and even has the potential for inventory depreciation. Ideally, the inventory level should be kept as low as possible. However, in reality, manufacturing companies face uncertainties from the market demand, production processes and supply. Advanced planning of production activities to hedge against these uncertainties is necessary and important for manufacturing companies. What is a reasonable amount of inventory to hold and how to manage the corresponding production planning and control have raised challenges to decision makers. Hence the aim of production and inventory control, in general, is to minimize the overall cost, including the inventory holding/backlogging cost and production cost, by determining the optimal control variables, such as production quantity, safety stock level and production speed. Although the problem of production and inventory control has been studied intensively, research gaps still exist and more practical factors should be taken into consideration. For example,stationary demand is always taken as an assumption in previous research works.However,in real cases, non-stationary demand can be commonly observed. Corresponding optimal control policies should be investigated.In addition, a large number of previous research studies normally assumed that the production system is perfect and the inventory is non-perishable, and the rest either only looked at problems with deteriorating production processes or perishable inventories. An integrated study considering both of the factors needs to be carried out. Hence this research aims to address the issues and has filled the research gaps mentioned above in the field of production and inventory control. There are five main deliverables provided in this research: First of all, non-stationary demand and forecasting have first been introduced into hedging point based production and inventory control policy. Two forecast corrected control policies have been proposed and optimized through an integrated simulation, design of experiment and response surface methods. Secondly, the impact of forecasting inaccuracy on hedging point based control policies has been investigated which can provide aid in the decision making process of choosing control policy. Thirdly, product and process deterioration have been jointly examined in two economic production quantity models together with backorder and rework. In addition,bothpiece-wise linear increasing and stochastic production process deterioration have been modeled.Optimality conditions for cost functions were obtained and numerical examples were employed to illustrate the performance of the proposed models. The optimum combination of production and backorder quantity in each production run were determined and sensitivity analysis showed the influence of different parameters on inventory behavior.