Concepts such as Just-in-Time (JIT) and Just-in-Sequence (JIS) have been applied in the automobile industry for many years; however, in many places, consistent implementation was lacking. Recently, new and promising approaches emerged based on AI-assisted software platforms. Then came the coronavirus. Now we need approaches that can help us manage global supply chains better.
First let’s take a look at the time before the coronavirus, when the world (and supply chains) still moved in familiar ways: Just-in-Time (JIT) and Just-in-Sequence (JIS) were well-known control concepts that were seldom applied completely and thus repeatedly revealed the gaps – or, to put things in more positive terms – the significant need for optimization in the automobile industry supply chain.
Of course there were also automatic planning processes in the most commonly used ERP systems, which frequently still had to be adjusted manually: from the comparison of incoming orders to availability of material to quality comparisons to detailed production planning. Excel was often used for these purposes. The weaknesses of this constellation have now been revealed in a painful manner by the coronavirus crisis, for global supply chains were suddenly interrupted due to many national governments’ lockdown orders.
It then became clear just how crucial transparency of suppliers down to the machine could be. The people in charge would like to have known, for example, when and whether the required materials would arrive. People who are lacking such important information cannot plan for the long term. Of course state-of-the-art SCM solutions couldn’t have prevented the breakdown of supply chains, however, they could have gained the people in charge valuable time to make decisions about alternative supply paths and sources.
If you use true JIT/JIS concepts as a benchmark for timely supply chain management, you can see that the possibilities extend way beyond transparency requirements. In particular, the goals of JIS and JIT address better coordination of ordering, planning, and production. Ideally, customer orders are transferred to production orders, these are then compared to the inventory in the supply chain, and this produces the requirements that are reported to the suppliers. These concepts should replace production that relies solely on forecast calculations.
Automobile manufacturers such as Porsche, BMW, and Daimler converted their supply chains years ago so that parts requirements are reported to suppliers at the last moment. The suppliers have to be in a position to deliver very quickly. This, in turn, requires an adjustment of the planning and control systems in the automobile suppliers’ production and logistics in order to adequately fulfill the new requirements of JIT/JIS concepts.
Basically, JIT/JIS concepts provide delivery of products or components directly to assembly lines. Generally this is only the case for large tier 1 system suppliers. An example is the automobile manufacturer Magna, which develops, produces, and assembles its components and systems for vehicle interiors in precisely timed specified sequences. thyssenkrupp also produces and delivers vehicle systems on customer demand, that is, in synch with the sequence of customer production.
Here too, the coronavirus revealed weaknesses, for JIT/JIS concepts allow for only +very small buffer inventories, which are only sufficient for a short period of production and thus allow for no errors in planning and control processes; otherwise, the assembly lines grind to a halt.
The lesson from the Corona crisis is that new, combined approaches that include digitalization and localization are required. A crucial modernization component is the introduction of software solutions that support continuous processes from suppliers to the machine. Such programs can help companies react very flexibly to demand fluctuations and OEM breakdowns, as well as to align processes in production and logistics with JIT/JIS concepts. Artificial intelligence (AI) methods can help automobile manufacturers and OEMS analyze structured and unstructured data so that they can forecast their needs more precisely. Unstructured data could be reactions from the Internet/social media to new vehicles or topics such as e-mobility. Conceivable is a better forecast of possible supply bottlenecks due to bad weather or other external factors such as the coronavirus. Structured data is the order history by customer group or inventory in the warehouse/the supply chain and reprocurement times.
A possibility for making supply chains more resilient could be shifting global supply chains to supply chains that are sooner regional/local. The coronavirus has demonstrated that the more global they are, the more vulnerable and opaque supply chains become. Newer developments are aimed at more local/regional and also more flexible approaches. These can include so-called “on-demand warehouse marketplaces” (e.g. LogHub, Stowga, Stockpots). These marketplaces offer available storage capacities of local and regional warehouse suppliers in order to provide manufacturers and retailers with greater warehouse flexibility and shorter delivery paths. For supply chain management, this creates another management level whose data has to be analyzed in order to satisfy the high requirements for transparency, resiliency, JIT/JIS planning, and forecasting and risk management.
The world of supply chain management is becoming more complex. Even before the coronavirus haunted global supply chains, weaknesses in manufacturing industry’s supply chains could be detected. Even in the automobile industry, where flows have been optimized and fine-tuned over the years, there was still room for improvement, on the part of both the OEMs and the suppliers. The coronavirus crisis has now mercilessly revealed these weaknesses. State-of-the-art, high-performance software and AI solutions can help confront these challenges.