Józefowska J. Just-in-Time Scheduling. Models and Algorithms for Computer and Manufacturing Systems

Springer, 2007. — 266 p.The philosophy of just-in-time manufacturing was first introduced by the Japanese automobile producer Toyota in 1950s. This philosophy may be briefly defined as elimination of waste and simultaneous continuous improvement of productivity. There are many different sources of waste in a manufacturing system; therefore, many activities need to be undertaken in a company in order to effectively implement the justin- time philosophy. Waiting time, overproduction and inventory are the sources of waste which can be eliminated by appropriate production planning and scheduling.
The goals of just-in-time scheduling differ from the goals considered in traditional production scheduling. Therefore new scheduling problems have been defined within the theory of scheduling to meet the need for practical solutions. Two optimization objectives are considered in the context of just-in-time scheduling. The first one is minimization of production variation, which means that the same amount of any output should be produced every day, or even every hour. An interesting aspect of the problem of minimizing the production variation is its similarity to the problem of apportionment. The results of the theory of apportionment are exploited in the analysis and design of scheduling algorithms used to minimize the production variation. The second objective examined in just-in-time scheduling is minimizing the total earliness and tardiness cost. Minimization of the total earliness and tardiness cost expresses the aim to reduce inventory cost and, simultaneously, satisfy the customer demands with timely delivery of products. This objective gives rise to non-regular performance measures, and thus leads to new methodological issues in the design of scheduling algorithms. Scheduling problems with both objective functions, i.e. minimization of the production variation and minimization of the earliness and tardiviii ness cost, which appear in just-in-time production planning and control systems, have found numerous applications in the control of computer systems. The most important class of computer systems working in a just-in-time environment is the class of the real-time systems. The main requirement for a real-time system is to respond to externally generated input stimuli within a finite and specified period. This requirement results in the same scheduling objectives as those considered in the justin- time manufacturing systems. Consequently, the same optimization algorithms may be applied to solve scheduling problems in just-in-time manufacturing systems and in real-time computer systems.
The aim of this book is to present both classes of scheduling problems and both application areas together, in order to show the similarities and differences of the approaches. The book contains a survey of exact and heuristic algorithms developed to solve the scheduling problems in the just-in-time environment. The presented survey may alert the reader to similarities of models and techniques used in different optimization domains, like those for the scheduling theory and the apportionment theory. Many concepts and algorithms are illustrated with examples, tables and figures to enhance the clarity of the presentation. As such, this book should differ from other surveys of just-in-time scheduling problems in the scope and the unified treatment of problem formulation and solution procedures.Just-in-time concept in manufacturing and computer systems
Methodological background
Common due date
Individual due dates
Algorithms for schedule balancing