Thesis (MComm)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: This study explore how forecasting techniques can be combined in linear programming (LP) as a tool to optimise the parameters of forecasting methods in order to ensure sufficient capacity of logistic infrastructure exist for future growth. This study will use greenfield and brownfield projects from Sasol, a petrochemical company from South Africa, to test the methodology on.
The methodology followed in the study was to firstly look at previous literature studies on logistical infrastructure and how to create sufficient capacity. Secondly, understandings of supply chain planning principles in general as well as supply chain planning in context of Sasol were investigated. Thirdly, different forecasting methods like; qualitative include judgemental, life cycle, Delphi method, market research etc. and quantitative methods including time series and causal methodologies had been investigated. Fourthly, decision making tools to incorporate multiple forecasts were investigated to understand why Sasol decided to use i2. Fifthly, the current capital project approach in Sasol had been investigated to fully understand where room for improvements would be possible. Finally the theory from the study was applied on two different projects in Sasol, one greenfield and one brownfield project. The results found that by using sound supply chain planning methodologies, sound supply chain design principles and multiple forecasts being combined by using LP decision making tools a better decision can be made with regards to logistical infrastructure investment as well as ensuring sufficient logistical infrastructure capacity. The two case studies have shown that this approach is flexible enough, apart from a few minor changes and can be adopted for both scenarios and that great results can be achieved. Logistical infrastructure could be optimised due to collaboration and the overall costs and performance of a supply chain improved. / AFRIKAANSE OPSOMMING: Hierdie studie ondersoek hoe lineêre programmering ( LP ), as n hulpmiddel, gebruik kan word om vooruitskattingstegnieke te kombineer om sodoende die vooruitskattingsmetodes te optimaliseer en te verseker dat voldoende kapasiteit van logistieke infrastruktuur bestaan vir toekomstige groei. Hierdie studie se metodes sal getoets word op groenveld- en bruinveldprojekte van Sasol , 'n petrochemiese maatskappy van Suid –Afrika.
Die metode gevolg tydens die studie, was eerstens om te kyk na vorige literatuurstudies oor logistieke infrastruktuur en hoe om voldoende kapasiteit te skep. Tweedens, om ‘n breë oorsig van die beginsels van voorsieningsketting-beplanning te bekom sowel as voorsieningsketting-beplanning in die konteks van Sasol te ondersoek. Derdens, verskillende vootuitskattingsmetodes soos kwalitatiewe metodes (insluitend veroordelende-, lewensiklus- en Delphi-metode en marknavorsing) en kwantitatiewe metodes (insluitend die tydreeks- en oorsaaklike metodes) is geondersoek. In die vierde plek is besluitnemingshulpmiddels, wat verskeie vooruitskattings kombineer, geondersoek om te verstaan waarom Sasol besluit het om i2 aan te koop. In die vyfde plek is die metode van Sasol se kapitaalprojekte geondersoek om te verstaan of daar nie moontlik ruimte vir verbeterings sou wees nie. Laastens is die studie se metode op twee projekte van Sasol toegepas, een groenveld- en een bruinveldprojek. In die studie is gevind dat beter besluite geneem kan word aangaande beleggings in logistieke infrastruktuur en om te verskere daar is voldoende logistieke infrastruktuur kapasiteit - deur gebruik te maak van optimale metodes in voorsieningsketting-beplanning en voorsieningskettingontwerp. Die twee gevallestudies het getoon dat hierdie benadering buigsaam genoeg is, afgesien van 'n paar klein veranderinge, om vir beide moontlikhede gebruik te kan word en goeie resultate te behaal. Deur die samewerking van verskeie besigheidseenhede kon logistieke infrastruktuur geoptimaliseer word terwyl die kostes en algehele prestasie van voorsieningsketting verbeter kon word.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/86539 |
| Date | 04 1900 |
| Creators | Gebhardt, Albertus Johannes |
| Contributors | Havenga, Jan, Stellenbosch University. Faculty of Economics and Management Sciences. Dept. of Logistics. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 101 p. : ill., maps |
| Rights | Stellenbosch University |
Page generated in 0.0024 seconds