A unified Data Model is the Answer
Do you need a bridge between your vehicle’s GPS system and its shock absorbers? Obviously, the answer is NO! Then why would you need to bridge planning and execution in the supply chain? This is a self-made and unnecessary problem when S&OP solutions plan inaccurately at a very high level without knowing what the true detailed constraints are. Therefore, resulting in inaccurate plans which would require somebody or some system to try and make it work! The shock absorbers in your car execute by constantly monitoring bumps on the road and taking care of the smoothness of the ride. If the ride gets to be too uncomfortable then GPS might kick in and change the route. Both GPS and the shock absorbers have their own function and level of intelligence relying on sensors to know when to “re-plan.” S&OP systems are incapable of doing this because they form inaccurate monthly plans which are not executable and creates inefficiencies in the supply chain. The feedback loop is non-existent in S&OP systems and the latency is too long. However, when you introduce S&OE then the feedback loop is closer to execution and closer to real-time. This helps, but it does not close the gap between S&OP and S&OE. To be able to solve this problem, it does not make sense to have high level planning system at the top, and low-level sequencing on the shop floor. The systems do not really talk to each other and have no idea what the other is doing.
In order to properly address this issue, one needs to have a unified data model that relies on the same modeling and logic from top to bottom. That is from network planning and S&OP all the way down to scheduling and sequencing. Thus, every bump on the road is responded to by the lower levels and when it gets too much the next level (scheduling) kicks in; and when that gets too far from the designed plan, then supply chain model takes over to redistribute the load and so on. With a unified data model there is no need for a “bridge.” The integration of planning and execution is already built into it. Furthermore, a unified data model, as described above, creates a continuous planning environment. Using the above analogy, you don’t stop the car to re-plan your route with GPS. As the problems appear on the road, traffic jams, or uncomfortable ride, the system keeps planning as needed. It also looks ahead to see if there are potential problems and how they can be avoided.
In supply chain planning, the look ahead is done by building accurate models of the supply chain (a Digital Mirror©). Using the Digital Mirror©, we can observe the issues ahead of time and deduce the potential patterns that are developing using machine learning techniques. As we plan, we are capable of predicting what potential risks are developing and what the causes are, e.g. relationships between inventory shortages, weather patterns, seasonal variations, and supplier uncertainty etc. As these patterns emerge, we can take actions accordingly using prescriptive techniques and optimization methods to avoid supply chain risks.