An Integrated Model For Optimising Manufacturing And Distribution Network Scheduling

Noor Ajian, Mohd. Lair (2008) An Integrated Model For Optimising Manufacturing And Distribution Network Scheduling. PhD thesis, University of South Australia .

[img] PDF (24 Pages)
An_Integrated_Model_For_Optimising_Manufacturing_and_Distribution_network_scheduling.pdf - Submitted Version
Restricted to Registered users only

Download (2MB)


Supply Chain Scheduling (SCS) emerged as a result of the in tegrated Suppl y Chain Management concept and hould be incorporated in planning and operating Supply Chain (SC). The review of literature indicated in tegration of SCS in the SC is still far from being achieved. Recogni s ing th at optimisation of scheduling activities separate ly will only lead to local optima, thi re earch propose a SCS global optimisation through the integration of scheduling within manufac turing and di stiibuti on networks. To address the need for integrating the manufacturing and dis tributio n networks. a centralised SCS framework has been proposed to handle the scheduling process. The centralised SCS system con ists of an optimisation system (the centralised cheduling system). the SC nodes and the system in puts and outputs. 1n addition. the integration of product and process flows with compone nts requirements is also introduced. to coordinate and handle the SC' huge et of data. Furthermore, a comprehens ive mathe mati cal model, which represents the proposed integrated SCS system, was also developed. The objective function is formed based on all re levant cost quality measu res. As a result. the mathematical model developed in this thes is is precise for evaluating the SCS performance. Genetic Algorithms (GA) have been ide ntified as a robust optimisation technique and have been selected for this research. However, the literature indicated variability in GA perfom1ance. Con eque ntl y. this re earch al o concentrated on improving the performance of GA for this type of problem by: developing a multi-dimens iona l representation to re present the complex SCS problem; using the Stochastic Universal Selection as the selection technique; and by hybridi sing the classical GA with a heuri stic. Comparative studies demonstrate that these improvements achieve a better and more con istent . et of o utcomes compared to o ther representations and operato rs. The proposed hybrid GA-based SCS optimisation system develo ped in this thesis has also been tested on a complex, industri al-s ize SC case study. Specifically, a three-eche lon Malaysian e lectronic SC with international markets has been successfully mode lled and schedul ed and cons is te nt and stable res ults were obtained. as compared to ex isting scheduling me thods. The success in scheduling of this SC and the good res ults indicate the feas ibility of the proposed centralised SCS framework and a lso demonstrate the effectiveness and effic ie ncy of the novel hybrid GA SCS optimisation system. Finall y, the performances of the proposed SCS global optimisation through the integrati on of manufacturing and distribution networks were evaluated by comparing the proposed integrated ne twork approach with exis ting approaches (the manufacturing and di stribution network approaches, respecti vely). The comparative analysis shows that both the SC schedule cost and varia bility for the integrated approach is s ignificantly lower th an the other two approaches. Specifically, the implementation of the integrated network approach leads to a reduction of at least 30% in the minimum SC total cost. 1n addition, remarkable system stability is also demonstrated by having the smallest standard deviation. These res ults highli ght that the proposed SCS integrated approach can achieve a prominent role across the SC.

Item Type: Thesis (PhD)
Uncontrolled Keywords: Production planning, Business logistics
Subjects: T Technology > T Technology (General)
T Technology > TS Manufactures
Divisions: Library > Tesis > FKP
Depositing User: Siti Syahirah Ab Rahim
Date Deposited: 02 Oct 2014 17:40
Last Modified: 28 May 2015 04:31
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item