Skip to content
Prescriptive Analytics

The Power of Prescriptive Analytics

IBM Decision Optimization is a prescriptive analytics solution. Prescriptive analytics helps organisations make better decisions by solving complex optimisation problems. Special software evaluates millions of combinations of information to give crystal clear recommendations. In other words, using decision optimisation gives you access to Oracle in the Matrix!

Decision Optimization works best when there are thousands of choices to make and resource limits. This forces the consideration of thousands of trade-offs. A human can only consider a small fraction of these choices and will usually rely on intuition and familiar patterns. In rapidly changing business landscapes, intuition and patterns are less effective than fact-based decisions offered by Decision Optimization.

In this three-part series we will:

1. Look at what Decision Optimization is and how it works

2. Share real-world examples of optimisation action

3. Help you identify optimisation applications for your organisation.

Let’s begin!

What is Decision Optimization?

Decision Optimization is a prescriptive analytics technology, leveraging several products such as CPLEX Optimization studio, IBM Decision Optimization for Watson, and IBM Decision Optimization Center to help decision makers handle the trade-offs that arise between options. Prescriptive analytics helps examine how you can achieve ideal outcomes for particular situations. And it can also help evaluate how to mitigate and avoid uncertain risks.

In the face of thousands (or millions!) of choices and limited resources, there are a few ways Optimization can help you:

1. Discover previously unknown options and approaches. This is because Optimisation can examine far more choices than a human decision maker could.

2. Automate or streamline decisions. This frees up your bandwidth, allowing you to focus on business challenges rather than routine tasks. An example includes personnel scheduling in retail, hospitality or any shift-based work environment.

3. Enable scenario analysis. This allows you to test your decisions under various contingencies. It can also give deeper insights into the trade-offs required to extract the greatest value with the limited resources available.

How Optimization Works

Optimization begins with an ‘optimization model’, a mathematical expression that specifies the relationships between targets, limits, and choices involved in the decisions.

You also have to input data, representing the resources available and demands to be met (e.g. number of staff, shifts, sales targets). Input data also includes costs, operational constraints, and business goals.

While input data can vary when analysing a problem, the optimization model doesn’t change. Let’s say you’re schedule planning. Whether you’re dealing with 100 staff or 10,000, the optimisation model stays the same. Having the model and data separated in Decision Optimization allows easy scalability.

The model plus the input data creates an ‘instance’ of an optimization problem. Optimization engines then use algorithms to find a set of decisions that best achieves the business goals required while respecting the limits imposed. The solution often gets summarised to the decision makers in tables and graphs that provide insight.

While Decision Optimization uses complex mathematical algorithms, you don’t have to understand this math to use it well. Consultants with expertise using the products in the Decision Optimization solution (Infocube being one of them), can take your input data, build optimization models and summarise solutions so you can stay focused on the decision making itself.

Subscribe to our newsletter to get the next post, which will look at specific examples of Decision Optimization in action.

Leave a Comment





This site uses Akismet to reduce spam. Learn how your comment data is processed.

Need help with TM1?
We're here for you

Categories

Tags

Tweet
Share
Email