Optimize the Colors in Your Cube

Create better strategies with operations research

By Mary Crissey

Imagine a Rubik's Cube with each turning block representing a unique business unit of operations. As decision makers strive for a bright future, they devise action plans to accomplish specific goals and to gain the best advantage in the hopes of winning the game with a solid color on each side. Real-world problem solvers know they're not just manipulating six sides of the cube with six different colors – but must in fact optimize hundreds of color combinations and thousands of different variables.

Spend too much time worrying about the blue side of the cube (inventory, perhaps), and suddenly the red side (the shipping schedule) is messed up. Turn to find a solution that puts all the red pieces in place, and the single row of yellow (purchasing) that used to be aligned is now three different colors.

Operations research (OR) makes it easier to solve the Rubik's Cube of your business operations. In particular, the latest release of SAS/OR software extends predictive power with optimization features so decision makers can analytically devise the best operational game plan. The updated software from SAS simplifies and streamlines the process so that organizations can implement proactive planning quickly and efficiently.

The history of operations research
Operations research is not a new concept. During World War II, the United States and Britain called on mathematicians to study ways to improve military operations. With a scientific approach, these OR pioneers simulated battle plan scenarios, identified troop requirements, planned logistical distributions and pinpointed the best actions to weaken the German army.

With so many different types of troops, military units and potential battlefields, you can imagine the immense number of possibilities in military planning. Operations research eliminates the need to test each and every possible combination by narrowing the choice of alternatives down to just the ones that best allocate troops and equipment aligned with military objectives.

After the war, some companies began to consider preplanning strategies; however, the mathematical, statistical and algorithmic functions required to examine and cross-reference operational processes were complex and difficult to understand and master. In addition, data collection was a tedious manual process, and because computer memory was both limited and costly, most organizations could not realize the benefits of operations research.

As organizations entered the information age and began to computerize operations, the focus shifted from the accumulation of data to business intelligence (BI). BI provided insight into operations with solid ways to measure progress by showing performance trends based on comparisons to history. Most companies now have gathered tremendous amounts of data, and leveraging that wealth of information has become a priority.

Putting company information to work
Operations research helps organizations rise above the reactive, busy mode of coping with daily challenges and guides them with proactive steps to create better action plans for success. OR allows organizations to leverage BI, get a glimpse of future scenarios and devise a "best" plan for operating business affairs so they can achieve enhanced growth and productivity.

Above all, optimization recommends decisive actions that collectively fulfill identified organizational requirements to ensure the most productive use of company resources. Resources may include people, materials, silos, plants, tools, machines or time. SAS/OR provides a scientific, logical and guided approach for organizations to follow as they consider strategies for optimizing business practices.

Today, operations research is being used to streamline and enhance operations across industries in a variety of concrete ways, including: 

• Resource allocation and management.
• Production and inventory planning.
• Product mix and blending. 
• Staffing allocations. 
• Distribution, routing, scheduling and traffic flow.
• Supply chain management and optimization. 
• Capital budgeting, asset allocation and portfolio selection.

Reducing complexity
Experienced modelers must translate business scenarios into mathematical representations before any software can perform the sophisticated programming techniques involved in operations research. Once the real-world problem is expressed in terms of constraints, goals and objectives, a wide range of decisions – including resource allocation, or product mix and blending problems – can be addressed and optimized.

With SAS/OR, the complex mathematical programming is packaged and complies with industry standards so users don't have to fuss with obtuse inputs or learn fancy lingo. With the SAS platform, decision makers can launch SAS/OR directly from their favorite spreadsheets. Those who use SAS Enterprise Miner™, SAS/STAT or SAS Forecast Server can integrate their analyses with the newly enhanced optimization product.

SAS/OR provides a single modeling language for many different types of problems. The software is smart enough to detect which type of solver, or computational algorithm, is appropriate for which business problem. As a result, users don't need to worry about declaring which mathematical technique to use (or know which niche vendor's software tool to buy). Often OR analysts are assigned to ad hoc "hot issues" teams, and with SAS, they can be confident that they are prepared to solve a wide range of optimization challenges.

Other enhancements make it easier to validate models, review them and make subsequent adjustments to better represent the real-world business cases and later rerun them with new data for similar situations.

In short, the analytics for operations research are complex. There's no way around that fact. And that's why it is important to select an OR application that is user-friendly and compatible with various BI sources.

With SAS/OR, you'll have a solid base for your analytical analyses as well as the flexibility to adapt easily to changing needs, and you'll have the optimization power to add insight, innovation and credibility to the decision-making process.

Bio: Mary Crissey is the Analytics Marketing Manager at SAS, where she follows her passion for applying mathematics and advanced analytics to real-world challenges.