Intelligent Logistics in the Automotive Aftermarket

Intelligent Logistics in the Automotive Aftermarket New tools and concepts for maximising profits through part specific optimising of • Stocking levels • Service levels • Supply and replenishment cycles • Order quantities • Customer rebate structures • Profit contribution 2 Background: Despite the continuing cost reduction efforts, shareholder value creation in the automotive industry has been constantly underperforming in comparison to other industries. In times of increased restructuring and globalisation however, capital has become the most critical resource. In their strive to increase shareholder returns, the automotive industry is rediscovering it’s most important asset: the customer base. With more than 250 Million first- hand customers and further 250 Million second-hand buyers globally, the automotive industry possesses the largest and wealthiest customer base of any industry. Unfortunately for the OEMs, the dealers often hold access to the customer base and therefore “own” the relationship, and not the OEMs. The OEMs do not only have to build bridges in order to reach the final customer. They also have to bridge the significant time gap between two car purchases of the individual customer. There are two solutions to this problem: 1. Using the Internet for gaining direct access to the customer and/or establishing company owned sales outlets 2. Selling more than just cars in order to bridge the time gap between two contacts In addition, the OEMs should get hold of the attractive second hand car market in order to maximise customer contacts and extend their potential after sales market as long as possible into the life of the car. Over the coming years, ever more additional products and services: telemetrics, in-car consumer electronics, software and hardware updates, lifestyle products and other associated sales items will be rolled out through the aftermarket channels of the industry. The traditionally strongest segment of the automotive aftermarket, the captive parts, could be pushed downwards from currently 75% to less than 50% of the total aftermarket business, making the market more volatile and prone to competitive attacks. Furthermore, the fall of the block exemption will accelerate competition in the aftermarket and put further barriers between the OEMs and their final customers. The traditional cascading logistics systems therefore need to be transformed from relatively closed, captive service units into open, highly competitive service networks. 3 Competition in and between these networks will primarily be carried out on the basis of product availability and service and to a lesser extent on the basis of price. Therefore those companies will be in the lead, which manage these factors best. Providing the best instant product availability and service level at lowest costs will be the most decisive and most competitive factor. Today, typically, 10% of captive part’s sales and 50% of competitive parts sales are lost to competition. There is great danger of losing further sales. Unfortunately, current instruments do not work well. High service levels are paid for with high inventories. Costly same-day or next-day emergency shipments compensate for stock shortages. Extensive parts tracing systems, collaborative stock locators and cross-network emergency supply arrangements shall cure the failures of these systems. What is often praised as the most advanced system is, in fact, a costly cure of the symptoms, not a cure of the problem. Our experience: Over many years, our partners have been dealing with the optimisation of numerous parts distribution networks across Europe and worldwide. Restructuring warehouse networks, establishing fast transportation links and providing 24-hour services to every dealer and repair shop have solved some of the issues of the past. At the same time stocking locations, inventories and costs have been cut. The most critical issue in these projects has always been the decision on the right level of inventory at each location. When setting and controlling the service policies and target inventories, real-world results often lagged behind the calculated optimum. We found out, that the reasons for this were a number of basic flaws in the inventory management systems, which were applied: 4 1. Demand forecasts never matched real demand 2. Safety stocks were not calculated but estimated. Often they were the result of a long- term trial and error process of the Inventory managers. 3. Targeted service levels were set at a flat rate for the whole stock instead of individual differentiation by part or by part group. 4. Tools were incapable of identifying optimal stocking policies by item, and managing and controlling the results. 5. The more refined forecasting models appeared to be, the worse were the results. Some items met target service levels, others not. In average, all analysed inventories showed that approximately 30% of items were understocked and 70% were overstocked. As a consequence, real inventories often exceeded calculated optimal stocks by 30% and more. In addition, the 30% of understocked items caused excessive use of emergency shipments, a larger workforce in the warehouse, more customer complaints and loss of sales. All this was not discovered by the normal tracking systems, which were applied. The reasons were, that the higher fill rates of the overstocked items compensated for the lower fill rates of the understocked items. The systems pretended that everything was well, whilst, in reality, it wasn’t. The identified improvement potentials were already impressive. However, parallel simulations of “should be” stock performance on the basis of real demand and real supplies showed, that our simplified calculation methods, in which we used actual demand variations of representative items as a measure to determine target stocks, still resulted in up to 10% higher target inventories than necessary. The reasons were found in the unpredictable random match between the random distributions of peak and trough demands and the random levels of real inventories at the given time. Our methods to determine target inventories may have been good enough when systems did not allow for better inventory management and when the biggest improvements could be made through stock consolidation and fast transport networks across Europe. It is clear to us, however, that current and future competition will demand more refined, individualised and more precise tools in order to achieve the optimum service levels. This not only has to be achieved in the central and regional depots of the EOMs but also throughout the whole supply chain, up to the last dealer and his associated stocking points and workshops. 5 The solution: In our solution, we have abandoned the idea of forecasting demands. Indeed, it has proved impossible to provide a realistic forecast for any demand between two economical replenishment cycles for any given individual item. And that is, what should have to be done in order to make current systems work. In reality, real per-item demands over short- or medium term forecasting periods are so erratic, that fixed period forecasts prove to be always wrong. In fact, they become even more wrong, the more the forecasting systems work with seemingly sophisticated trend-following methods and the like. Current forecasting and ordering systems also do not take into account the real flexibility and responsiveness of today’s production set-ups and therefore do not exploit their potential in full.