What’s the problem? In a Supply Chain Dive opinion article published in 2020, Gartner’s Christian Titze offered his critique on the subject, saying that “control towers are like the artificial intelligence of supply chain: everyone wants to have it, but nobody quite knows how it works.”

This seems to be the perfect breeding ground for a confused market with many half-baked solutions going live, offering little more than a data warehouse with BI reporting. In the best-case scenarios, these reports signal critical deviations occurring in the supply chain, pinpointing the causes. In the worst cases, they signal either too much or nothing at all. What most of them fail to do is provide actionable information.

• Stage 1: Detect - Any supply or demand event with a significant positive or negative business impact is presented clearly and at the speed of the business (meaning not always in real time but as close to real time as the business requires). The reporting should cover every level of the supply chain, from strategic to operational and from sourcing to delivery. It should also be configurable to allow companies to finetune the solution so that only the critical issues are signaled.
• Stage 2: Diagnose - The potential bottom-line impact of any signaled event (for instance on revenue, service level, or costs) can be calculated and presented, along with the probability of occurrence and its reach within the supply chain.
• Stage 3: Respond - Scenarios can be defined and benchmarked against each other, allowing the most viable option, in terms of current and future business impact, to be selected. The solution should also make it possible to integrate automated response scenarios and the conditions under which they will be triggered.
• Stage 4: Learn - A continuous improvement loop means that lessons are learned from the events to be integrated into the solution, gradually enriching the decision-making process with automated and semi-automated solutions.