Sunday 10:45 AM–11:30 AM in Speakeasy

Prescriptive Analytics with DOcplex and pandas

Veronique Blanchard

Audience level:
Intermediate

Description

This talk provides an overview of prescriptive analytics technology and illustrates how DOcplex, an open source Python package, supports the development of optimization models. Relying on pandas and the Python ecosystem, a complete prescriptive analytics workflow, from data import and pre-processing to visualization of optimization results, can be implemented in a single language

Abstract

Prescriptive analytics technology recommends actions based on desired outcomes, taking into account specific scenarios, resources, and knowledge of past and current events. This insight can help organizations make better decisions and have greater control of business outcomes. Prescriptive analytics used to rely on dedicated modelling languages like AMPL, GAMS, OPL… DOcplex is an open source Python package that intends to provide the same flavor as those dedicated languages, both in terms of expressiveness and performance, thanks to pandas. It can actually run with community edition (free) of IBM CPLEX, a free for academics edition or a commercial edition. This talk provides an overview of Prescriptive Analytics and illustrates with examples how DOcplex, combined with tools available in the python ecosystem, can smoothly support the full workflow from data import and preparation, to visualization of optimization results.