Design of a Warehouse and Distribution System for Imported Beers

Abstract

The aim of this project is to provide a new company, African Beer Importers, with a set of designs for their new warehouse in Silver Lakes, Pretoria. The designs were tested using a simulation model and compared according to several metrics; the most prominent being the Total Handling time, which measures how much time is spent handling the pallets, and the second being the Rack Utilisation, which measures how much the racks are being used. Two solutions excelled, one for each metric, and so both are being presented in order for the company to make a nal decision based on whether they prioritise space or time more. The rst design uses a Chevron-Aisle oor pattern, with a 3-Class storage policy. The Chevron aisles minimise the travel distance incurred on the reach trucks, by approximating the Euclidean distance (as the crow ies)from the o oading point to any particular pallet, and minimising the tedious and traditional rectilinear movement pattern. The 3-Class storage policy allocates positions on the shelves based on the popularity of the beers, and puts the most frequently collected beers closest to the loading area, and the least used beers at the back, so that the most common trips are the shortest ones. The second design uses a Leaf-Aisle oor pattern, with a Random storage policy. The Leaf pattern allows for more storage slots than the Chevron pattern, but has shorter travel distances than the traditional long, horizontal aisle patterns. The Random storage policy is actually pseudo-random, as the beers are simply stored in the rst available position. This allows for maximum shelf-space utilisation, but comes at the cost of long material handling times and confusion in fi nding the correct beers.