Prescriptive Analytics

Descriptive Analytics analyses past data to understand trends present in the data to answer what happened. Predictive Analytics techniques will predict what will happen to key performance indicators in the future. Prescriptive Analytics assists the decision maker to identify the best action (optimal solution), given the problem context. That is, prescriptive analytics, as the name suggests, prescribes the best solution or decision/action for the problem. Note that decisions or actions can be derived based on descriptive and predictive analytics as well. For example, using predictive analytics, a retailer such as Amazon and Flipkart can predict what a customer is likely to buy in the future and design product recommendations. The difference between decisions arrived at using descriptive/predictive analytics and prescriptive analytics is that the prescriptive analytics algorithm tries to arrive at the best decision (optimal solution) based on an objective function (sometimes more than one objective function) and a list of constraints.Operations research techniques such as linear programming, integer programming, goal programming, non-linear programming, and meta-heuristics are used for prescribing an optimal solution to a problem. A few big data problems originated from optimization problems. For example, the travelling salesman problem (TSP) is one of the most difficult problems which is encountered by organizations such as online retailers, logistic service providers, and even electronic parts manufacturers. Prescriptive analytics models attempt to solve complex optimization problems of the modern era.

Week 1: Prescriptive Analytics and Linear Programming (LP)

Week 2 – Sensitivity Analysis and Duality

Week 3 – LP Formulations and Applications

Week 4 – Integer Programming (IP)

Week 5 – Multicriteria Optimization

Click here for syllabus page

Exam schedule