By Marcin Figlarek
Just 10 years ago, in Poland the statement went around that the main task of a WMS system is electronic recording of events in the warehouse, i.e., generally working with bar codes. Of course, many warehouse systems on the market back then operated in just that way and it seemed to be sufficient. Warehouse employees worked at terminals, but really, they or the warehouse manager needed to decide what to do and when. Selecting the appropriate process at a terminal, the employee had to sometimes decide which release document to choose in order to start working. Actually, it is hard to call IT systems that operate in such a manner Warehouse Management Systems. A more adequate name for such solutions is a record or registration system. For many small warehouses employing only a few employees, such a system most likely will be sufficient, while in the case of larger warehouses, there is still a large opportunity to optimize.
A WMS provides control and transparency
Advanced Warehouse Management Systems operate differently. The warehouse employee does not decide which release document to choose. In fact, the employee doesn't know for which customer the goods are for. Ultimately, what would be the purpose of such information? The task of the warehouse employee is effective, flawless work and performing the assigned tasks at the time assigned by the WMS system. Using the task approach it is possible to use many types of picking organization processes. An example is the ability to split orders into many work areas, such as a separate area for picking on the mezzanine floor and a picking area from the pallet racking. On the other hand, picking planning processes, so-called arranging, can be single stage (at order level) or multistage (at items level). It is also possible to accumulate orders, that is, the possibility of picking goods from various orders in one picking cycle. When working on documents, as opposed to working on tasks, it is not possible to choose between so many different picking strategies. Optimization usually ends with determining the most optimal path for the collection of goods.
Work on tasks does not exhaust the reach of WMS system capabilities. The system can efficiently assign tasks to employees, but it still does not fully make use of the available time for all employees. Each of them works within a given process. The employee who puts the pallets on racks will be working on moving tasks, the person picking the goods on the tasks of picking, while the forklift operator which replenishes the goods from the stock to the picking places will not be given other tasks than replenishing tasks. In theory, it would appear that such work is optimal, while statistics show something completely different. It turns out that 50-60% of the work time in the warehouse is so-called empty runs, i.e. ineffective work. Something is done but it doesn't bring any benefits. An example of this is the fact that most transport cycles of forklifts are performed with empty forks. Naturally, a question arises: is it possible to minimize these “empty kilometers”?
Increase warehouse efficiency with 40%
The answer is a system such as Astro WMS, which can perform so-called task interleaving. The warehouse employee who puts a pallet from the docks to the storage area, on the way back for another pallet from the receipt area, can deliver a different pallet by putting it to the docks area. On the other hand, the employee who replenishes the picking area from the storage area can perform on occasion a maintenance tasks. This is determined by the WMS system based on the priorities of tasks assigned automatically and then creates, so-called multicycles. Thanks to this approach, according to several independent studies, you can obtain an increase in warehouse productivity from 10 up to 40%. For medium-sized and larger warehouses these are huge savings in the total cost of manufacturing and delivering the product, where logistics costs account for few to a dozen or more percent.
Picking planning in WMS systems varies in terms of parameterization. Some WMS systems require processes strictly “tailored” to the customer's requirements, while others can be configured in a way to enable future changes by the system’s user, which translates in lower system operation costs (less custom-made modifications). Which picking stratregy should be selected? What if it proves to be inaccurate? How will the WMS system perform during peaks?
The answer for full-pallet releases are so-called smart tasks, meaning adaptive task interleaving available in the advanced module of the Astro WMS. It consists of a dynamic picking selection strategy depending on, among others, where exactly the warehouse employees are located, what the task priorities are, as well as the total transport time of goods. How does the WMS system estimate these times? Well, estimations are difficult to be made, at least at the beginning of implementation, so the system is “self-learning”. It collects on a regular basis all the time data for moving goods between various points and creates a "warehouse time map". As a result, the WMS system knows what pallet to select for release, so that the execution time would be the fastest. First, it searches for goods listed for release according to the shipping priority. Then it checks whether there are no equivalents of these pallets in the working corridor, in which the forklift operator is already located. If the goods are there, it decides on the release of the goods in which the total picking time is the smallest. If there are none, it selects the goods originally designated and checks the times of total movements making the best selection. This method of picking brings double-digit productivity growth, according to Consafe Logistics clients.
On the other hand, in the case of pieces picking, Astro WMS can be operated in either release priority or minimum path priority, where this mode is determined automatically by the WMS system based on the available time and amount of orders. When there is little time and the highest priority orders should be collected as quickly as possible, the WMS system designates for the employee locations for the collection of goods urgently required. In this case, the forklift can travel long distances, because the priority is the collection of goods urgently required. On the other hand, when we have more time, the WMS system can switch over to the mode for minimizing trips. This consists of the WMS system first searching for the locations of goods most required. Then in the vicinity of these locations it searches for other goods with less priority. The goal is to visit as many locations as possible in a small area and "filling the fork".
This is certainly not the end of innovative picking methods. Modern WMS systems are racing to create new algorithms, which would not have been possible to implement using paper and pencil. An increasingly important role will be played by systems like Astro WMS, which can self-learn and improve itself, minimizing not only operating costs but also the time required to configure and administer the WMS system.
Figure. Within the area designated by locations with the highest priority goods (red) Astro WMS searches for locations of goods with less priority (other colours) to "fill the fork" of the forklift operator. Source: Consafe Logistics.