טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentKirill Karmi
SubjectCombined Job Scheduling and Tool Requirement Planning for
Flexible Manufacturing Systems
DepartmentDepartment of Industrial Engineering and Management
Supervisor Mr. Rubinovitz Jacob (Deceased)


Abstract

The last decade was characterized by the increasing technological breakthroughs in all production spheres. The increasing trend towards tailoring products to meet customers needs has led to a further reduction in manufacturing batch quantities, an increase in product varieties and shorter product life. One of the major developments in metal removing industry was the introduction of the flexible manufacturing system (FMS) concept that has emerged as a viable answer to the problems of flexibility and efficiency. However, proper resource distribution strategy together with the jobs and tools assignment policy have become a key condition to the system’s success in dealing with flexibility and efficiency issues.

This thesis deals with two main logistics-oriented FMS sub-systems: job scheduling and tool management systems and their interaction. To date, the impact of job scheduling policy on the amount of tools used during a production plan execution has hardly been recognized. However, without the analysis of this influence there are no accurate measures available to calculate tool lifetime needed for the given production plan execution and no true system production cost minimization can be performed. This thesis resolves the above problem through development of the mathematical model that takes into account planning rules for job scheduling, which are based on a set of real production environment assumptions. Several new concepts are developed for better and more realistic design of the production environment suitable for flexible manufacturing systems. The objective function is defined as minimum system cost engaged for production plan execution during a predefined planning horizon, subjected to applicable logistic restrictions. The genetic algorithm approach is proposed here as the most suitable for this complex non-linear problem. Schedule results for five sets of data support the validity of the proposed model and the practicality of its use in the metal removing industry.