Improving operations at a fulfilment operation at PostNL E-Commerce Services by improving the order volume forecasts and controlling planning parameters:
Institution: | Delft University of Technology |
---|---|
Department: | |
Year: | 2015 |
Keywords: | fulfilment; e-commerce; planning; forecasting |
Record ID: | 1242418 |
Full text PDF: | http://resolver.tudelft.nl/uuid:d45c9ad5-6b4a-42b2-91cf-17bb307e8f8d |
PostNL E-Commerce Services is an e-fulfilment centre, which means that they keep stock, pick, package and send out goods for web shops. For a web shop they handle the European distribution. On this packaging operation it turned out that 33% of orders were handled late. A discrepancy was found between the amount of orders placed and handled every day possibly resulting in late orders and increased cost. To reduce this discrepancy a simulation model was created to test the impact of different planning decisions on the performance. By experimenting with different forecasting models, the forecasting accuracy was greatly increased. However, the effect on the amount of late orders was minimal. Other planning parameters that were researched were the schedule resulting from the forecast, the starting time s and the maximal shift duration d. From these parameters, starting time s turned out to have the greatest impact. By changing the planning either late orders can be reduced from 33% to 13% of total handled orders or cost could be reduced by 9%. The full results are mapped and can be used as a guidance for planning on this operation.