When I entered the production manager's office, he was already waving a stack of papers. I thought to myself "whoops, what does this mean?". It was a 1 cm thick computer printout of all the delayed production orders. That really sounded bad.
After a brief preliminary discussion, I set to work to understand what the problem was. The production facility had over 2000 different end products and had fallen behind due to a glitch in the program, with no production orders being transmitted for a month.
After analyzing the production and order data, the following critical areas emerged, creating a vicious circle:
🩺 Despite superhuman efforts of the production staff, the backlog did not improve, but getting steadily worse every day.
🩺 Many customers were already out-of-stock for a number of their products.
🩺 The production schedule was constantly being changed as customers called top management to have their own orders prioritized. As a result, product sequencing was no longer optimal and changeover times accounted for an overwhelming share of the available capacity.
After discussing various alternative solutions, we decided on the following approach:
🩹 The first thing to do was to reduce the excessive amount of changeover time. To achieve this, we agreed on a production schedule that was strictly optimized with regard to changeover times. In addition, order sizes were doubled, in particular to avoid machine changeovers for small quantities.
🩹 Decisions regarding the production plan were aligned with the crisis steering committee. It was agreed that the production plan would be frozen and that there would be no rush orders or other top-down interventions during the crisis.
🩹 In addition, the development of production capacity, changeover times and order processing was tracked in parallel as well as the order backlog.
As a result of these activities, we were able to gain 30% more capacity. On this basis, it was possible to give customers a transparent indication of when normal deliveries could be expected again - on the one hand, providing light at the end of the tunnel and, on the other, making the situation at least somewhat predictable for customers. The backlog, which had already lasted several months before the start of the project, could be worked through within two months, so that normal operations could be resumed.
What were the lessons learned from this backlog crisis?
💡 One should not blindly trust automated ERP and MES systems - rather, one should perform plausibility checks from time to time.
💡 Even the best automated systems only work if the master data is complete, realistic and error-free.
💡 Top-down interventions rarely help solve a problem. Rather, it is about bringing everyone to the table and seeking a common solution, committing to it, and then acting on it.
After solving the specific problem of the backlog and the supply crisis, the project naturally continued with other improvement measures such as de-bottlenecking, reduction of change-over times and better integration of the planning processes with the customer demand.
On a personal level, I am always particularly motivated by such crisis intervention projects. While the situation can be very stressful, I find it very rewarding to see an organization come out of crisis mode when the problem is solved.