Search results

1 – 3 of 3
To view the access options for this content please click here
Article
Publication date: 17 February 2012

Roongrat Pisuchpen

In the real world many companies combine the operations of manufacturing, assembly and disassembly. Thus, the integration of just‐in‐time FMS, FAS, and flexible…

Abstract

Purpose

In the real world many companies combine the operations of manufacturing, assembly and disassembly. Thus, the integration of just‐in‐time FMS, FAS, and flexible disassembly system (DAS) models poses an interesting problem. The purpose of this paper is to provide major emphasis on a new simulation model for design and performance evaluation of a flexible assembly and disassembly system with dual Kanban under a stochastic system. This paper also primarily investigates the effect of varying the number of kanban cards, mean inter‐arrival time of demand and locations of the bottlenecks on the performance integration of JIT flexible manufacturing, assembly and disassembly systems.

Design/methodology/approach

Simulation is carried out in ARENA and data is analyzed using multivariate analysis of variance (MANOVA). This paper investigates the effect of varying number of kanban cards, mean inter‐arrival time of demand, and locations of the bottlenecks on the performance integration of JIT flexible manufacturing, assembly and disassembly systems. The performance measures that are simultaneously considered are the fill rate, work in process, and mean cycle time. This paper emphasizes that understanding the interactions between the variables and their effects on system performance is of utmost importance for managers in improving performance processes.

Findings

In manufacturing practice, there are many industrial units that represent the mixture of the referred three models. This paper presents a new simulation model for design and performance evaluation of a flexible assembly and disassembly system with dual kanban. The simulation results are statistically compared with MANOVA. MANOVA is used to perform the test with multiple objective functions, e.g. with the average production cycle time, percentage average fill rate, and work‐in‐process. The conclusion to be drawn is that minimized WIP can be obtained by higher percentage average fill rate, lower WIP, small average part cycles times, and increasing in kanban cards while simultaneously retaining full customer satisfaction.

Originality/value

The researcher presents the newly developed kanban system into the production system of JIT flexible manufacturing, assembly and disassembly system with simulation technique. Furthermore, by assigning time factors to the models, several performance measures can be easily computed. Then, the researcher tests the effect of the number of kanban card on integration of JIT flexible manufacturing, assembly and disassembly systems using a simulation approach, the simulation model is developed using the ARENA simulation package. The results are applied to a small case study. For a single product under the integration of JIT flexible manufacturing, assembly and disassembly systems, as the number of kanban cards increase, the fill rate along with work in process and the mean cycle time increases as well.

Details

Assembly Automation, vol. 32 no. 1
Type: Research Article
ISSN: 0144-5154

Keywords

To view the access options for this content please click here
Article
Publication date: 12 April 2011

Shaukat Ali Shah, Erik L.J. Bohez and Roongrat Pisuchpen

A major factor to the success of flexible manufacturing systems (FMSs) is their ability to transport work pieces between different workstations. FMS have now become more…

Abstract

Purpose

A major factor to the success of flexible manufacturing systems (FMSs) is their ability to transport work pieces between different workstations. FMS have now become more advanced and material‐handling systems have become progressively more sophisticated, it is not exceptional to have automated steering of tools to workstations as well. Such system design will improve the tool‐handling capability and the system productivity while holding tool cost to a minimum. Tool cost could represent as much as 25 percent of the operating cost. The purpose of this paper is to propose a new colored Petri net (CPN)‐based approach to the design and development of a tool sharing control system that is intended to help use of the proper and minimal number of tools for a manufacturing system.

Design/methodology/approach

A new black token timed PN model is first developed, to reduce the complexity of the graphical representation a new CPN model is developed. The new CPN model also allows to find the optimal sequence. The optimal sequence has no effect on the work in process (WIP) but it influences the number of tools used in the system. The main input to the PN model for a manufacturing system is the process plan. Next, all the invariants and total number of possible elementary circuits are determined using the Integrated Net Analyzer (INA) software. Output from the INA software is exported to the Excel spreadsheet. The Excel spreadsheet can be designed to calculate the total number of tokens, processing time, cycle time, etc. of each elementary circuit. Subsequently, the constraints used in Lingo will be created according to critical circuit rules. Finally, linear programming (LP) technique is used to optimize the WIP and tool inventory. Lingo software is used for the LP, the constraints from the Excel sheet will be the input data to the Lingo program, and based on those constraints the Lingo will provide the optimal values for the desired parameters. The output from Lingo will be used to recalculate the cycle time of each elementary circuit in the Excel sheet. The system is then analyzed before and after the implementation of the CPN model.

Findings

A new CPN model based on tool‐sharing philosophy for an FMS with N part types and M stations is proposed.

Originality/value

The paper presents a new CPN‐based approach to the design and development of a tool sharing control system, that is, intended to help use of the proper and minimal number of tools for a manufacturing system. The new CPN model also allows to find the optimal sequence. The idea is new and pure and has not been presented before using the methodology adopted in this paper.

Details

Assembly Automation, vol. 31 no. 2
Type: Research Article
ISSN: 0144-5154

Keywords

To view the access options for this content please click here
Article
Publication date: 26 July 2013

Asawin Wongwiwat, Erik L.J. Bohez and Roongrat Pisuchpen

The purpose of this paper is to propose a new generic hybrid Petri Net (PN) model combined with the lowest makespan cut (LMC) for job shop scheduling problems in mold…

Abstract

Purpose

The purpose of this paper is to propose a new generic hybrid Petri Net (PN) model combined with the lowest makespan cut (LMC) for job shop scheduling problems in mold manufacturing to minimize the makespan of the mold part manufacture schedule.

Design/methodology/approach

The LMC algorithm finds a solution close to the optimal solution. The searching of the LMC algorithm starts from the lowest estimated makespan (lowest makespan). Almost all of the lowest makespans (LM) are infeasible makespans. A shifting percentage (SP) is added to the LM to obtain the shifting makespan (SM). The SM is compared with the completion time computed from the reachability tree of the Petri Net (PN) model. If the completion time is greater than the SM, the corresponding branch is cut from the reachability graph, and the SM will be compared with another branch from the reachability tree. There are two scenarios. In the first scenario, there is no feasible solution resulting from the comparison of the completion time and the SM, because the SM is lower than all of the feasible solutions. Therefore, the SP is used to increase the SM. On the contrary, in the second scenario, there is a feasible solution: the SP is used to reduce the SM. In the first scenario, a makespan that is lower than the optimal makespan is found. In the second scenario, a makespan that is greater than the optimal makespan is found. After getting close to bounds of the optimal makespan, the least makespan found in the bounds is the best solution.

Findings

The integration of the Petri Net (PN) model and the LMC algorithm can help to improve the production efficiency. In a case study, the proposed algorithm is being compared with other heuristical methods which are practical examples of mold makespans based on the shortest and the longest processing times. The schedule or the sequence obtained by the proposed algorithm is 30% less than the other methods.

Research limitations/implications

This research will consider scheduling multiple mold. The mold design and the mold testing phase are not considered.

Practical implications

The time to produce a mold is very important. Reducing the mold production time will provide more time for mold assembly and testing. The aim of LMC algorithm is minimize the makespan. The time to produce a mold is reduced by finding the best sequence of the jobs and machines.

Originality/value

This paper proposes the new generic hybrid Petri Net model combined with LMC for job shop scheduling problem in the case of mold making shop to optimize the makespan of mold parts scheduling.

Details

Assembly Automation, vol. 33 no. 3
Type: Research Article
ISSN: 0144-5154

Keywords

1 – 3 of 3