Artificial Intelligence (AI) is present in our everyday lives, from digital voice assistants on our smartphones or home appliances, to driving an almost autonomous car. We often do not even realise that there is AI behind the systems we use.
Artificial Intelligence obviously plays a role in our professional lives as well. Planning systems, for example, are increasingly powered by AI. They have become highly advanced systems that take into account a vast number of variables and constraints. Too many for the human brain to cope with. Until recently, the complexity of these systems was unimaginable.
The advantage of such advanced planning systems is that your demand plans, your forecasts, become more accurate, and your supply chain plans become more reliable. Which makes your supply chain more efficient and more effective.
But there is also another side to the coin. The more complex the AI-based planning systems are, the more difficult it is for humans – for the planners - to understand the output that is generated by the system, and to accept the decisions that are proposed by the system. The lack of transparency of how the algorithms transform the input data into demand and supply chain plans, makes the system work like a black box.
Now, why would this be a problem? Well, if we perceive a system as a black box, our trust in the system will suffer. Why would the planner trust the plan generated by an AI system if he or she does not understand the system’s reasoning to get there? What if the system’s recommendation is wrong or biased? And if the planner does not trust the system, he or she will find a way to bypass the system.
Explainable artificial intelligence is specifically designed to prevent this black box phenomenon from happening. Explainable AI consists of a set of tools and techniques that help humans not only to understand but also to interpret the recommendations that were made by the AI model. The purpose of Explainable AI is to increase the overall transparency of the system. It helps to make the shift from an opaque black box toward a crystal-clear glass box. This way, explainable artificial intelligence also leads to an increased trust in the system.