FAQ 3: What is prescriptive analytics
In this series you will find answers to questions we get asked a lot when speaking about data, analytics, digital transformation, emissions reduction, industry 4.0, IoT and data-driven automation.
What is prescriptive analytics? How it can help operations?
Prescriptive analytics is the next step after predictive analytics. It is a set of optimisation techniques that uses past and present data to give recommendations on how to improve the outcomes. In operations we do not just want to know what will happen to the outcomes, but what inputs can we change of optimise against to always get the best possible outcomes. The output of prescriptive analytics at any given time is a list of prescriptions designed to achieve the best result. Example applications for prescriptive analytics are divided between two areas:
-
Production planning, scheduling and sequencing. These problems are solved using mixed integer programming math and application of powerful optimisers like CPLEX and Gurobi, or robust heuristics.
-
Process and product optimisation. These problems are solved using a combination of machine learning to build reliable input-output models for process outcomes, and constrained optimisation techniques to optimise these models in real time when conditions change.
