Dynamic Production Scheduling Research Articles

Dynamic scheduling for complex engineer-to-order products

The current paper considers dynamic production scheduling for manufacturing systems producing products with deep and complex product structures and complicated process routings. It is assumed that manufacturing and assembly processing times are deterministic. Dynamic scheduling problems may be either incremental (where the schedule for incoming orders does not affect the schedule for existing orders) or regenerative (where a new schedule is produced for both new and existing orders). In both situations, a common objective is to minimize total costs (the sum of work-in-progress holding costs, product earliness and tardiness costs). In this research, heuristic and evolutionary-strategy-based methods have been developed to solve incremental and regenerative scheduling problems. Case studies using industrial data from a company that produces complex products in low volume demonstrate the effectiveness of the methods. Evolution strategy (ES) provides better results than the heuristic method, but this is at the expense of significantly longer computation times. It was found that performing regenerative planning is better than incremental planning when there is high interaction between the new orders and the existing orders.

2503 段取り時間を考慮した動的生産スケジューリングシステム(S77-1 生産システムの新展開-基礎(1),S77 生産システムの新展開(基礎理論を中心に))

2503 段取り時間を考慮した動的生産スケジューリングシステム(S77-1 生産システムの新展開-基礎(1),S77 生産システムの新展開(基礎理論を中心に))

An integrated agent-based approach for responsive control of manufacturing resources

Due to the dynamic fluctuation of customer demands in the global market, manufacturing enterprises are facing difficulties in rapidly responding to market changes. The aim of this research is to develop a system to integrate dynamic process planning and dynamic production scheduling for the purpose of increasing the responsiveness of adaptive manufacturing systems in accommodating dynamic market changes (rapidly changing demand patterns or product varieties). The concept of Multi-Agent Systems (MAS) has been adopted in this study. All of the tasks related to process planning, optimization and scheduling in this system are carried out by autonomous agents that are capable of interacting and negotiating with each other to bid for jobs and make decisions. On the one hand, this system optimizes the utilization of manufacturing resources and on the other hand, it also provides a platform where the reconfiguration of manufacturing systems can be assessed. This system has been implemented on a Java platform and a case study is provided to elaborate on this system and evaluate its implementation.

A dynamic reactive scheduling mechanism for responding to changes of production orders and manufacturing resources

This research introduces a dynamic reactive production scheduling mechanism for modifying the originally created schedules when these schedules cannot be completed due to changes of production orders and manufacturing resources. Production order changes include removal of an order that is canceled by a customer and insertion of an order that has to be completed within a short period of time. Manufacturing resource changes include breakdowns of machines and sudden sickness of workers. Match-up and agent-based collaboration approaches are employed to modify only part of the originally created schedules for improving the reactive scheduling efficiency, while maintaining the scheduling quality. The dynamic reactive production scheduling system was implemented using Smalltalk, an object oriented programming language.

Dynamic Production Scheduling for a Process Industry

Capacity-oriented production scheduling can be described as the assignment of competing products to several single level, capacitated production lines over a given planning horizon. This study was initially motivated by the production planning of various types of tiles by a tile manufacturing company. We considered different integer programming formulations and found that a disaggregated model, while increasing the size of the model, lends itself best to Lagrangian techniques and produces the strongest bounds. Additionally, from every Lagrangian solution, we can generate a feasible production schedule by systematically reassigning lines from products whose production exceeded demand to products with unsatisfied demands. The technique is not specific to tile companies, but can be used by any firm where product setups on production lines can be scheduled between consecutive periods with changeover cost, but without production loss. Our computational experience with real data from the tile company and randomly generated problem instances gave excellent lower and upper bounds.

Studying the effects of production loss due to setup in dynamic production scheduling

We study the effects of production loss during setup in dynamic production scheduling for process industries producing several products on non-identical flexible processors. The production scheduling problem is formulated as a mixed integer programming model and solved using primal and dual Lagrangean based procedures. Real data from a tile company and randomly generated data were used to test the model. Our computational results show the solution procedure to be fast, robust and stable even for cases with large setup costs and production losses, and for cases with demands of up to 52 periods.

Decision Support System for Dynamic Job—shop Scheduling

ABSTRACT This paper deals with dynamic production scheduling in a discrete manufacturing environment. Current MRPII methods do not allow scheduling to be handled easily while alternative, e.g. ‘pull’ methods, such as Kanban, require a stable environment—contrary to the flexibility demanded by rapid market changes. A new approach is therefore required to support production and schedule planners in their day-to-day activities. The approach presented here was developed in the scope of the European projects ESPRIT and EUREKA. It combines the expertise of the planner with the data gathering and processing ability of the computer in order to simulate planning strategies and ultimately obtain a good workplan. A decision support system based on a software package called Interactive Cell Integrated Management (ICIM) illustrates the approach presented here with an industrial implementation.

Applying a modified coefficient reduction method to a dynamic production scheduling model

This paper develops and tests a short-run production scheduling model for process industry application. We consider a multi-product, multi-period environment, where a subset of the products can be produced on each of the production lines at a fixed rate per period. The resulting 0–1 mixed integer programming problem is difficult to solve using commercial optimization software. A preprocessing method based on coefficient reduction was modified to exploit the problem structure. This preprocessing tightens the problem prior to solution by a commercial branch-and-bound code. The remaining duality gap is defined as the percentage difference between the best feasible value and the tightest bound found, i.e. it is the maximum improvement in objective function value that can still be expected. Gaps of less than 4% were found in all test problems when production losses were zero. With the inclusion of production losses, preprocessing led to about a 50% reduction in the remaining gaps. Additional computational studies found that this approach can process constraints having up to 80–110 variables. Further research is suggested to investigate the computational effectiveness of other solution approaches.

Decomposition principles applied to the dynamic production and work-force scheduling problem

One of the most important problems in the production and inventory planning field, is the scheduling of production and work force in a dynamic environment. Although this problem can be formulated as a linear program, it is often quite difficult to solve directly, due to its large scale. Instead, it might be fruitful to use a decomposition approach. Decomposition, in general, means decomposing a difficult problem into several easier, or a sequence of easier problems which are later coordinated to reconstruct the original problem. In this report we discuss several possibilities of applying the most common decomposition principles, namely Benders (primal) and Datnzig-Wolfe (dual), as well as a relatively new decomposition method, called cross decomposition, to the dynamic, multiproduct production and employment planning problem. A number of special cases are also presented.

A dynamic production scheduling model with lost-sales or backlogging

Two production scheduling models with lost-sales, backlogging and concave costs, are considered. The associated production and inventory levels of these models are considered to be bounded in every period. The structure of the optimal production schedule under these assumptions is characterized. In addition, efficient backward dynamic programming algorithms are suggested for both the backlogging and lost-sales cases.