Supply Chain Design & Controlling
To reach the best possible efficiency level in a multiple-order-picking process the right combination of orders is a key success factor. This multiple-order-picking process sets the minimization of the picking energy as the primary goal. On the other hand, the restrictions like departure time have to be strictly kept.
The principles of multiple-order-picking processes are very different in the industry. The main challenges are not the diversity but the high amount of order lines, the complexity of individual restrictions and the shortest possible time limits for the calculation of sequences with the help of the algorithms.
This leads to bad solutions when trying to find a solution manually in a time critical situation. The optimal process is crucial for the efficiency, particularly in such a situation, as many order positions have to be rescheduled and processed within a very concise time frame.
Similar findings are also known from the processes where the orders of the following day are processed during a night shift. In this case huge amounts of orders have to be planned within minutes. Common warehouse management systems (WMS) use a batch approach. Articles are being summarized into a batch and within the batch picking process the final orders are being generated.
A greater benefit can be achieved if optimization technology comes into play during batch creation. WMS can execute this only when they are equipped with special algorithms that can deal with so called NP hard optimization tasks (these NP hard optimization tasks typically lead to an exponential increase of calculation time when the number of variables is growing). In this area combinatorial algorithms have showed the best results. They allow customer specific amendments with respect to many industry related restrictions and heuristics and can be individualized. On the other hand, these algorithms can be tuned to find fast and efficient optimal solutions (like known from Branch-and-Bound or Branch-and-Cut solutions).
The graphic below shows by example how different orders with articles form different areas can be assigned to routes. Depending on the time frame and the dead line, orders can be changed between batches.
Potentials in a range of two digit percentage can be achieved either in systems with ten- or hundred-thousands of order lines by means of interactive sequencing systems and within a very short time needed for each recalculation.
Read also about
© 2017 Logivations GmbH. All rights reserved