Ebook Production scheduling: Part 2

Part 2 book "Production scheduling" includes content: Cyclic production scheduling, hoist scheduling problem; shop scheduling with multiple resources, open shop scheduling; scheduling under flexible constraints and uncertain data - the fuzzy approach; real time workshop scheduling.
Nội dung trích xuất từ tài liệu:
Ebook Production scheduling: Part 2 Chapter 7 Cyclic Production Scheduling 7.1. Introduction The cyclic character of the rhythm of life is the most natural of things. The most logical cause is linked to natural environmental phenomena which humans have to face: the inevitable alternation of seasons due to the Earth’s rotation around the Sun, as well as daily and nightly rotations due to the revolution of the Earth around its axis. Another cause is linked to traditions and society planning work based on a week containing work and rest phases. In summary, everyone lives according to a cyclic organization and this way of life naturally inspires different production management methodologies. In order to simplify and facilitate understanding of this presentation, two examples will illustrate this methodology. The first example, developed in this introduction, has no other objective than to present how activity management can be very complex and how this complexity can be sensibly reduced via cyclic organization. The second example will involve flexible manufacturing systems (FMS). It will be used as explicit support for this presentation. The introduction example involves a known, albeit complex, problem: class schedule planning within an institution (college or high school). The idea is to organize and plan an annual timetable of classes in an institution. Four types of entities gravitate around this problem: teachers, classes, classrooms and class schedules. The goal is to make sure students can attend classes corresponding to their curriculum in appropriate classrooms in the presence of a teacher specialized in the field involved. Chapter written by Jean-Claude GENTINA, Ouajdi KORBAA and Hervé CAMUS. 168 Production Scheduling Numerous capacity constraints involve the number of teachers per discipline, and the number and capacity of classrooms as well as the fact that only one teacher is assigned to teach in a given room at a given moment for students of a given class. As for the courses themselves, several characteristics must be taken into consideration. There are different course types with different timelines: classes, classroom studies (CS) and practical studies (PS). Teaching sessions are grouped into variable size modules depending on certain constraints; typically these constraints are precedence type constraints: classes before CS before PS. The first and foremost objectives involve feasibility of planning over a maximum possible timeframe (school year). Once this constraint is resolved, it then becomes possible to attempt to close the cycle as soon as possible in order to leave as much time as possible for students before finals, all the while respecting program timing constraints for all disciplines. Another reason justifying this criterion comes from the fact that it is always important to leave the largest margin possible at the end of the year to react and adjust the timetable for unexpected events (absence of a teacher, etc.). A second criterion consists of reducing the call for substitute teachers to minimize overheads. The compromise will favor student satisfaction because the establishment is concerned about their well being and success. The image commonly associated with this type of problem involves a large table where the person responsible for the timetable works very hard inserting little cards with different colors. This job is particularly difficult because he must model in two dimensions a problem which is by definition four dimensional: classes, programs, teachers and classrooms. The solutions generally used consist of splitting this problem into sub- problems (distribution by semester for example), then for each semester, a typical week is created. This typical week will be repeated throughout this period. The major complexity with the scheduling problem for resources (teachers and classrooms) is thus reduced to a temporal horizon; of a semester and a week respectively. This distribution of the semester load comes in fact from a first phase called short term planning. The goal of scheduling, logically called cyclic, thus consists of organizing the week load to be repeated over a semester as best as possible. In this chapter, we will focus on this way of organizing production activities by attempting to repeat a basic cycle relatively well optimized, thus illustrating the notion of cyclic scheduling. The advantage of such an approach, notably the finite optimization of this basic cycle, would remain useless without a certain control of the transition from the planning and scheduling levels. In fact, the periods retained must be carefully chosen at planning level whenever possible to represent the best compromise between the generally antagonistic feasibility of the solution retained, complexity of the approach (planning and scheduling phases) as well as optimizati ...