Dynamically Adjusting the Number of Kanbans in a Just-in-Time Production System Using Estimated Values of Leadtime

This paper evaluates several artificial intelligence heuristics for a simultaneous Kanban controlling and scheduling on a flexible Kanban system. The objective of the problem is to minimise a total production cost that includes due date penalty, inventory, and machining costs. We show that the simultaneous Kanban controlling and scheduling is critical in minimising the total production cost (approximately 30% cost reduction over scheduling without a Kanban controlling). To identify the most effective search method for the simultaneous Kanban controlling and scheduling, we evaluated widely known artificial intelligence heuristics: genetic algorithm, simulated annealing, tabu search, and neighbourhood search. Computational results show that the tabu search performs the best in terms of solution quality. The tabu search also requires a much less computational time than the genetic algorithm and the simulated annealing. To further improve the solution quality and computational time for a simultaneous Kanban controlling and scheduling on a flexible Kanban system, we developed a two-stage tabu search. At the beginning of the tabu search process, an initial solution is constructed by utilising the customer due date information given by a decision support system. The two-stage tabu search performs better than the tabu search with a randomly generated initial solution in both the solution quality and computational time across all problem sizes. The difference in the solution quality is more pronounced at the early stages of the search.

Section: A Brief Literature Review: Kanban Controlling and Schedulingmentioning

confidence: 97%

Evaluating artificial intelligence heuristics for a flexible Kanban system: simultaneous Kanban controlling and scheduling

Lee

2007

“…With constant interarrival times for demand, a single kanban is adequate. The majority of research has concentrated on determining the number of kanbans when replenishment lead time is variable but demand is static (Monden 1981, Rees et al 1987, So and Pinault 1988, Deleersnyder et al 1989, Philipoom et al 1990and Wang and Wang 1990. Mitra and Mitrani (1990) address a similar situation but assume an inexhaustible raw material and demand and compare it to finite buffer tandem queueing models.…”

Section: J 174mentioning

confidence: 99%

Production planning for a multi-item, single-stage kanban system

Mitwasi

Askin

1994

The kanban production authorization system has been widely implemented as a control scheme for just-in-time manufacturing. This system is normally applied in systems with reliable processes,low setup times,static demand and excesscapacity. The use of kanbans avoids the need for complex information and hierarchical control systems on the shop floor. Kanban control also has the advantage of reacting quickly to minor demand changes. In this paper we investigate the use of kanban control at workcentres which produce multiple items with dynamic, random demand. The dynamic aspects of demand may cause temporary capacity shortages. The kanban control system automatically reacts quickly to the random aspects of demand. Selection of the number of kanbans will accommodate the dynamic aspect. We assume an environment with low setup times and a desire for rapid adjustment in production rates to match demand. A mathematical model for the problem is defined and shown to reduce to a simpler problem. Using the simpler problem, we investigate necessary and sufficient conditions for feasibility of the problem and specific choices of kanban levels. These conditions are used to define upper and lower bounds on the optimal number of kanbans which, in turn, are employed for developing optimal and near-optimal, heuristic algorithms for selecting kanban levels for each item. An algorithm is also derived for finding feasible production schedules given a feasible set of kanbans. The algorithm provides a policy which is easily implemented on the shop floor for guiding production decisions when capacity is limited. The methodology is illustrated with an example and results of experimental testing of the algorithm and heuristic are presented.

“…Taking this into account, Rees et al (1987) and Gupta and Al-Turki (1997) proposed a flexible Kanban system that dynamically adjusts the number of Kanbans and the corresponding buffer size in response to the uncertainty in processing time and demand. However, they did not consider unstable changes in demand.…”

Section: Introductionmentioning

confidence: 99%

Comparing reactive Kanban systems

Takahashi

2003

In an effort to achieve agile control in a just-in-time (JIT) environment, reactive JIT ordering systems have been proposed for multistage production systems with unstable changes in demand. In reactive JIT ordering systems, the time series data for the demand are monitored, and unstable changes in demand are detected using statistical control charts. Then, in these systems, the number of Kanbans and the buffer size at each stage are adjusted as a response to the detected unstable changes. Recently, an adaptive Kanban system based on inventory levels was proposed and applied to unstable demand conditions. However, its performance has not yet been analysed and compared with the other reactive system. Therefore, the first purpose of this paper is to analyse and compare the performance of the reactive Kanban systems, both the control chart-based and the inventory-based systems. In addition to comparing their performance, as the second purpose, a new inventory-based system is proposed. In the proposed system, the time series data for the demand is not monitored, but instead the inventory level of a product is monitored to detect unstable changes in demand, and the number of Kanbans is adjusted as a response to these detected unstable changes. The performance of the proposed system is investigated and compared with those of the previous systems by using simulation experiments that model unstable demand conditions.