| | 9 NOVEMBER 2024between production steps, pick and pack times in dispatch, good receipts times, as well as setup times for tools and cycle times for production. All this data is summarized under the umbrella of Master Data. As a key enhancer for business success, Master Data control is still one of the big game-changing factors. In a poorly controlled master data world, the employees day by day have to react to misleading planning information from the system. For example, if a timing parameter is wrong, or the weight of a product, or a cycle time, there are in the real business, two chances to detect them. Either the experienced employee knows about it and tries to manipulate already in his mind with "Brainware" the system instead of using the "software" intelligence. This leads to a higher workload for the master planning employees. Or, the wrong parameter is pushed through the system into the planning cycle and leads to wrong scheduling assumptions, for example, for the duration of a production run cycle. This is blocking capacity on the machines that might be used for a different material production. Consequently, following this wrong planning approach, you will have a deviation in the output performance of produced parts. In many cases, this weakness is hidden by creating more material inventory, which means increasing the working capital. Or, in the worst case, it will lead to a shortage of products, which can only be solved by accelerating and expediting the delivery of products to the next customer in your chain. In another case, the weakness of a poorly managed planning cycle is shown in overtime of production, which means for night shifts or weekend shifts. Even if, per standard calculation, the whole installed machine capacity would fit the relevant customer demands.A straightforward master data control landscape can avoid a lot of mistakes done this way and improve the cost factor for administration people, as well as OPEX for solving the problem afterward. In general, for all data controlling units, there is a payoff resulting in fewer errors than before in the MRP run. The main weakness at the moment is that many companies try to solve the data quality issue with manual controlling from business experts or so-called subject matter experts to clean the data in the system manually. Unfortunately, already small companies are challenged with 50k to 100k or more material numbers ­ multiplied by 20 to 30 or more of the most important master data fields, depending on the chosen ERP system. This kind of manual control is going to fail. Therefore, step 2 of data controlling must be integrated into the mass data cleaning process, with the defined logic from the business matter experts. Your IT department can create data cleaning tools for getting better accuracy in this area. What must be followed up in future data-cleaning business cases is the usage of AI for running this kind of report. In the beginning, the AI system will just help to identify the logic breaks of the data fields (which can easily be multiplied in millions of data in the Material ~ and Master Data Matrix). At the same time, with repeated learning, AI tools might be able to multiply and speed up the way of sustainable data quality improvements. This leads finally to fewer assumptions from the "brainware" of the process experts people and will help to minimize planning cycle errors because of wrong assumed data. In general, for all data controlling units, there is a payoff resulting in fewer errors than before in the MRP run
< Page 8 | Page 10 >