Ch. 2
Gaining Competitive Advantage with DSS

Chapter Contents Previous Page Next Page

The following examples at Frito-Lay, L.L. Bean, Lockheed-Georgia, and Mrs. Field’s Cookies should clarify how Decision Support Systems can provide a competitive advantage. These examples are "classics" that have been widely reported in business case studies and the popular press.

The Economic Optimization Model system required a shift in focus for the company from a traditional service-level criterion, such as 14 calls per agent per hour, to a method that would optimize economic efficiencies. This model-driven DSS examined variables such as the number of telephone lines to carry incoming traffic, the number of agents, and the queue capacity or the number of wait positions for sales agents. Then, through various mathematical modeling tools, the system generates specific resource amounts L.L. Bean should deploy to be most economically advantageous. The system takes into account many variables. For example, installation and maintenance costs of telephone lines, labor costs of sales agents including their training, costs associated with being on hold with the 800-service provider, and the cost of permanently lost orders. This new profitability-based model continues to add resources until the marginal cost of additional resources exceeds the return on that investment. EOM also scheduled the resources based on fluctuations in activity. When one group of operators became overwhelmed the next shift of operators would be starting and then as things were becoming slow one group of operators would soon be leaving.

From a $40,000 capital investment in the system, the company estimated a profit gain of $9.2 million to $10 million for 1989. Sales call volumes were up 6.5% over the previous year, managers attributed the majority of the gains to the new EOM system. Managers benefited from an integrated planner that could evaluate "what if" scenarios. Most importantly L.L. Bean’s reputation with customers improved. Other benefits included decreased customer wait times, improved morale of employees, and lost order penalties reduced. The EOM provided L.L. Bean a competitive advantage.

In 1975, Robert B. Ormsby, President of Lockheed-Georgia, a subsidiary of cargo aircraft producer Lockheed Corporation, was interested in the development of an online reporting system that could provide top executives with concise, timely, relevant information that could be shared within the organization to aid with decision making. The goals of the new system would be the insufficiencies of the existing system. In the fall of 1978, development began for a Management Information and Decision Support (MIDS) system (cf., Houdeshel and Watson, 1987).

The intended benefits of MIDS were improved communications, an evolving understanding of information requirements by the organization, and cost reductions in the generation of reports and presentation materials. MIDS helped managers identify areas that require attention; thus enabling improved decision making. Information has become more timely since it is updated as events occur and accuracy is improved through the verification of all information before it is made available.

Eventually a Knowledge-Driven or Suggestion DSS was developed that automated routine activities and responded to exceptions by prompting the store manager for input. Eventually these exceptions were structured and the system responded automatically to many situations.

On a more sophisticated level, the system tracked financial performance of each store, provided comprehensive scheduling of operations, including marketing support, hourly sales goals, and even assisted with candidate interview selection. Each store’s Tandy PC accessed the corporate management system. Many applications were menu-driven such as, day planning, time clocks, store accounting, inventory management, interviewing schedules, skill testing, and e-mail. After entering basic workday characteristics the system would run a mathematical model to compute the day’s schedule of events including how many cookies to bake of each type, when to mix and cook them and projected sales per hour. As store sales were periodically entered into the system then revised projections and recommendations would be made. With sales and inventory information the system prepares and generates supply orders. Headquarters was able to learn quickly when a store was not meeting expectations and managers could immediately respond.