Universiteit Stellenbosch
Welkom by Universiteit Stellenbosch
Intreerede deur prof Stephan Visagie
Outeur: Ronel Beukes
Gepubliseer: 08/05/2018

​Prof Stephan Visagie, professor in Operasionele Navorsing en departementele voorsitter van die Departement Logistiek, het sy intreerede op 3 Mei gelewer.

Hy is voorgestel deur prof Ronel du Preez, visedekaan: leer en onderrig van die Fakulteit Ekonomiese en Bestuurswetenskappe. Sy het gesê prof Visagie se navorsingsbelangstellings sluit onder meer die volgende in: nie-lineêre knapsakprobleme en die toepassing van optimeringstegnieke soos heeltallige programmering en metaheuristieke in diverse velde soos landbou, produksie, logistiek, skedulering en stemteorie.

In sy lesing oor optimering in distribusiesentrums het prof Visagie die gebruik van distribusiesentrums in die voorsieningkettings van kleinhandelaars toegelig. Dié sentrums dien as voorraadbuffer en word ook gebruik om die voorraad wat in groot maat van verskaffers ontvang word, in kleiner versendings te verpak vir verspreiding aan winkels.

Vir verdere besonderhede oor die inhoud van sy lesing, kyk hier onder.

Prof Visagie het al talle toekennnings ontvang, waaronder drie rektorstoekennings vir buitengewone prestasie, en verskeie eerbewyse van die Operasionele Navorsingsvereniging van Suid-Afrika, onder meer vir “'n uitstaande bydrae tot die toepassing van Operasionele Navorsing op 'n nasionale vlak" vir die werk wat hy saam met Pep Stores gedoen het. Die vereniging het hom boonop met sy hoogste toekenning vereer deur hom 'n genoot van die Operasionele Navorsingsvereniging van Suid Afrika te maak.

  • Foto: Prof Stephan Visagie by prof Eugene Cloete, viserektor: navorsing, innovasie en nagraadse studies, en prof Ingrid Woolard, dekaan van die Fakulteit Ekonomiese en Bestuurswetenskappe.


In his lecture entitled Optimisation in distribution centres, Prof Visagie explained the use of distribution centres in the supply chains of retailers to buffer stock and to repack the bulk stock that arrive from suppliers to smaller shipments destined for stores.

“The retailer's central planning office releases batches of stock-keeping units that differ only in size. These batches are called distributions and are continuously released to the distribution centre to pick and ship to stores. A distribution thus contains one or more stock-keeping units together with instructions about the quantity to be shipped to individual stores. Each distribution also has a deadline, by which date it should have left the distribution centre.

“The system for the repacking of stock – or order picking – consists of parallel unidirectional picking lines. Order picking occurs in waves, when a set of distributions is assigned (and physically taken) to a specific picking line. A team of pickers then pick all the stock-keeping units on this particular picking line for all the stores. Once all the stock-keeping units are picked, the picking line is populated with a new set of distributions to start a new wave of picking."

According to Prof Visagie this methodology resulted in three layers of optimisation decisions, while the objectives considered during these decisions were to minimise the following: walking distance of the pickers, the volume of stock on the picking line, the number of packages of which the volume fail to exceed roughly the size of a shoebox, and the total penalties incurred to distributions for not being assigned to a picking line before a specified date.

“All proposed solution methods were tested on real historical data received from PEP Stores Limited. The preferred combination of algorithms – genetic algorithm for first tier, user-specified in the second tier and greedy-ending heuristic in the third tier – showed an average decrease of 41% in the walking distance of all the pickers and a 34% decrease in the volume of stock on the picking line with the largest volume of stock. The percentage of small packages was maintained while on average a 20% decrease in the average number of days that distributions are assigned later than their out-of-distribution centre-date was achieved.