Plug-in optimization : the last mile

From an idea or a paper to some optimization benefits : the last mile.

In order to deploy and benefit from optimization there are many options:

• Custom optimization: Write a model and connect that model to an existing or a new application
• Tailored optimization: Use a high level platform in order to build a tailored optimization application without starting from zero

• Package: Use an off the shelf application that embeds optimization

Those are pretty standard. And we all have them in mind most of the time.

But let's not forget other options where we do not start a new application but add optimization to a tool that is there already. The additional cost and risk are?often smaller than the three first options. The fit may be a good enough compromise between a package and a tailored application sometimes.

Consider any data repository as data you may use for optimization and prescriptive analytics:

• Excel
• Database
• Maximo
• Matlab
• SAP HANA
• IBM Cloud Pak For Data

Or even add optimization on top of

• descriptive analytics

1. Introduction video with CPLEX in Planning Analytics / TM1

1. 5+ ways to connect PA and CPLEX ( DO-TM1_Connector, TM1py, WatsonStudio, Through Excel SheetWrite / SheetRead , TurboIntegrator)

• predictive analytics / Machine Learning / Data science

1. SPSS Modeler
2. IBM Watson Studio data science platforms to help data scientists do more with less too

More generally plug-in optimization may add value to any tool using an external call. And with CPLEX in the Cloud no need to instal CPLEX on your server or machine to plug in optimization!