Modeling and Optimization for Short-term Scheduling of Plastic Bag Plants

Abstract

This work presents a mixed-integer nonlinear programming (MINLP) formulation for the short-term scheduling of a multistage multiproduct batch facility producing different types of plastic bags. Given a number of customer orders that must be satisfied at specific due dates, the proposed model is able to determine the optimal assignment of orders to units and the sequence of orders on each unit at every processing stage. In contrast to previous works, it allows calculating both the processing rate of each equipment and the processing time of each product as a function of the bag?s parameters. Also, the number and diameter of film rolls obtained in the extrusion step as well as the product quality are accounted for in the model. Two alternative performance criteria are used. The first minimizes the delay in the delivery (tardiness) of all orders in the scheduling horizon while the second minimizes the total time required for processing all the orders (makespan). The model efficiency has been evaluated using real data from a plant located in the city of Guayaquil, Ecuador. The results demonstrate the improvements obtained with the proposed model when compared to the actual situation in the company.