IBM ILOG CPLEX Optimization Studio vs. SAP Predictive Analytics

In my opinon, if the problem is less than 5000 variables, one should try to solve with free available solver rather than directly going for a commercial license of IBM CPLEX Optimization Studio. In my opinion, if the priority is not in terms of solving time with higher number of variables, even then one can go for free solvers like CBC, IPOPT, SCIP. In my opinion, if priority is solving time and number of variables is also high, only in that case one should prefer going for a commercial license.

When you don't have much data science skills or when you need to build something quick, you can get started really fast with SAP Predictive Analytics. There are scenarios, like the ones covered by a neural network, where SAP Predictive Analytics is not well suited, but the big question is usually: "do you really need a neural network to address this use case?"

Incentivized

• Linear Programming
• Mixed-Integer Linear Programming
• Non-Linear Convex-Optimization
• Visualization
• Shadow Price Analysis
• Parameter Tuning

• Ability to use built-in algorithms or expand using R. This means that (with training) casual users can take advantage but also data analysts can do their thing!
• Integration and consistency with Lumira. Even Lumira on its own has a quick 'predict' functionality (although limited/black box)
• Ability to do the analysis and then present visually using the Lumira visualization capabilities

Incentivized

• Data handling from different sources like Note Pad, etc.
• Large size of MILP problems.
• Various parameters to set.

• The "User Experience" is sometimes lacking some clear basic things. Maybe a migration to a cloud-based environment will help bridge that gap.
• API is probably the next item on my list. The existing one is not easy to access or use which limits the integration capabilities.

Incentivized

It's nice to use and with good optimization.

the UI is a bit dated and available as a desktop tool mostly.

Incentivized

Honestly, to say, I never contacted CPLEX but used its forum to know/clarify any issues I faced.

The documentation provides an explanation about what features are available but not necessarily what's happening behind the scenes. On the other side, the "community" has grown since the acquisition and most questions are properly addressed by SAP folks. Since the "product maintenance" mode announcement was made, there wasn't much new content published except on the Smart Predict side (which is built by the SAP Predictive Analytics team)

Incentivized

Compared with MATLAB, CPLEX is a more user-friendly and simpler structure for writing models. This one also has a good return on investment.

We have typically used Spotfire for data analysis but decided to move to SAP Business Objects due to its innate connection with SAP. I found Lumira to be good for visualizations but it is not meant for data analysis. Therefore, we have introduced Predictive Analytics to see if it can fill that gap. So far, it's been far less intuitive than Spotfire to get started, and as far as I am aware so far, it does not bring many additional capabilities. I do, however, like that it utilizes the Lumira look/feel and integrates very well.

Incentivized

• Better for price.
• Many model parameters/features.
• No visualization.

• Proper forecasting increases our credibility with partners and customers
• Forecasting determines the amount of investment in each sector and reduces the cost of additional costs

Incentivized