An investigation into efficient multiple command order picking in high bay narrow aisle warehouses
With the increasing tempo of business, order picking efficiency becomes of increasing importance. The aim of this research is to find ways of increasing order picking efficiency by decreasing travel time of the picking cycle. Finding the optimal (shortest) order picking tour can be equated to the task of finding a solution to the Travelling Salesman Problem (TSP). This is notoriously difficult to solve in reasonable time when conventional computers are used. A number of heuristic algorithms have been developed for solving the TSP, some of these have been specially adapted for multiple command order picking. In this previous work the stacker crane's shuttle was assumed to travel with constant velocities in the horizontal and vertical directions. In this research it is shown that this assumption leads to creating suboptimal tours. The contribution of the error is analytically derived and its magnitude estimated by a simulation experiment. In previous work it has been shown that the shape of the zones in class based storage affects the travel time for single and dual command order picking. In this research, for the first time, the interaction between class based storage and multiple command order picking is investigated. Three types of zone configurations are modelled and then investigated using simulation in a factorial experiment. The results from the experiment indicate that the zone shape does affect the optimal solution. The new zone configurations are tested in a case study against existing configurations in a distribution warehouse of Volkswagen - Audi (VAG-UK). This showed that overall improvement in travel time of the new configurations was significant. Computer simulation was used to estimate the individual contribution from zoning and tour construction.