Planning and scheduling in supply chain environment within process industry

Process manufacturing is common in the food, beverage, chemical, pharmaceutical and consumer packaged goods industries. There is a continuous stream of input materials and output products. The structural characteristics of the process industry are sequence dependent setup times, high changeover costs, numerous flavored and colored product types with the complicated changeover rules, limited shelf life restricting the storage duration and delivery conditions for each perishable raw material, intermediate and final product. In this environment, efficient planning and scheduling of the supply chain is of vital importance and has become one of the most challenging problems in practice. This dissertation concerns planning and scheduling problems in the process industry. The main goal is to develop mathematical formulations of the supply chain problems. In this thesis, the characteristics of the process industry are analyzed. The major trends and research opportunities are explored from the existing literature. The existing models in the literature do not address many realistic aspects of the planning and scheduling problems in the process industry. Starting from this point of view, a production and distribution problem is studied in the soft drink industry. A mixed-integer linear programming model is introduced and, due to the high complexity in production and distribution structure, a hybrid solution methodology is developed to solve the realistic problems. The operational scheduling represents the realization of tactical planning decisions in operational level. Having created a plan for defining the production tasks has to be sequenced ensures that the planning activities are indeed applicable. The integration of planning and scheduling can be an effective way to make more applicable decisions. Accordingly, a production and distribution problem is studied in the dairy industry. A mixed-integer linear programming formulation is introduced to integrate tactical planning and operational scheduling decisions and, a heuristic approach is proposed to decompose the different time buckets of the decisions. In real life, intermediates are more perishable than final products. The final products can survive for long shelf life periods, but the lifetimes of the intermediates are only restricted with several hours. The perishability should not be only included in inventory level or shelf life of final products, it should also be realized that the perishability limits the intermediate storage and affects run-lengths of production. A scheduling problem is studied in the make-and-pack production process. A stochastic mixed-integer linear programming model is introduced to schedule the production. A simulation of the production process is introduced to evaluate the proposed production schedule in terms of the production waste, mostly caused by the variability in lifetime of intermediates. In summary, the industry specific characteristics, incorporation of the decision levels providing interrelated feedbacks to each other and perishability issues are recent challenges confronted by the process industry. These challenges require specific models to support decision making in supply chain. In response, this thesis develops mathematical models and optimization approaches applicable to different processes industries and can easily be modified for process specific operating conditions.