Six Sigma Analysis Improves Facilities Management Processes at NCI

Quality Program Enhances More Than 1.4 Million sf of Research Space
Published 5-17-2005
  • New Facility Construction

    SAIC-Frederick Inc. uses the Six Sigma approach to determine the best way to proceed with construction, design, and renovation projects at the National Cancer Institute in Frederick, Md. Here, tilt-up walls are being set in place for the new 125,000-sf CGMP facility. Photo courtesy of SAIC-Frederick Inc.

    Photo courtesy of SAIC-Frederick Inc.

  • Laboratory Renovation

    This new lab at the NCI's Center for Cancer Research contains a tissue-culture room, dark room, and various work and equipment rooms. The rooms are outfitted with Fisher Hamilton casework and phenolic resin countertops and shelving. Photo courtesy of SAIC-Frederick Inc.

    Photo courtesy of SAIC-Frederick Inc.

More than half of the 110 buildings located on the 68-acre campus of the National Cancer Institute (NCI) in Frederick, Md., are at least a half century old. Sixty-seven of the buildings date back to the 1940s and 1950s, while only 13 have been built since 1990. A campus-wide revitalization, estimated to cost between $1 billion and $3 billion, is expected to occur within the next five to 10 years. In the meantime, the aging structures present facilities management challenges for SAIC-Frederick Inc., a subsidiary of Science Applications International Corp. Until new buildings are constructed to meet modern standards, SAIC-Frederick continues to find ways to maximize efficiency of the existing structures.

"Older facilities were often constructed to now out-of-date codes and this problem is exacerbated by the fact that only two buildings have complete as-builts," says Dante Tedaldi, deputy director of facilities, maintenance, and engineering at SAIC-Frederick. "This can lead to unknown conditions that appear during project execution and impact cost and schedule by requiring redesign, new materials, or other work changes."

NCI's Frederick center, part of the National Institutes of Health (NIH), is located within Fort Detrick, a U.S. Army base. NCI-Frederick focuses on research designed to identify the causes of cancer, AIDS, and related diseases. More than 100 principal investigators are researching the environmental, behavioral, genetic, and molecular factors that contribute to human cancers and AIDS, as well as identifying new targets for diagnosis, treatment, and prevention.

The total staff of government and contract employees on campus is approximately 3,000 with about 1,500 of these individuals working directly through SAIC, which has an annual operating budget in excess of $320 million for the Frederick site. More than 40 percent of the 1.4 million gross sf of building space is used for research support with 26 percent being used for laboratories, 21 percent for administration, and the remainder for animal holding facilities.

SAIC Organization and Services

The facilities management organization of SAIC consists of a director who is responsible for the immediate supervision of operations and maintenance, project controls and administration, engineering, and project management/construction. About $20 million is devoted to the facilities management budget each year.

The annual workload typically includes about 9,000 trouble calls, hundreds of planned work orders, 3,000 planned and preventive maintenance tasks, 4,000 special assists on projects less than $5,000, and 80,000 hours of janitorial services. SAIC emphasizes safety on all of its projects and has seen a dramatic decline in the number of OSHA-reportable accidents, dropping from more than 30 in 2001 to less than 10 in 2004.

"Typically, we do most of the design and construction for projects up to several hundred thousand dollars and sometimes up to $1 million. We subcontract for the larger projects," says Tedaldi.

About 75 percent of the work completed by SAIC is valued at less than $50,000 and includes projects such as modifying labs or adding counter space. There are exceptions to the rule, however, in terms of the magnitude of projects handled by SAIC, which is currently completing a $65-million job to design and construct a vaccine pilot plant for in-house drug manufacturing for clinical trials.

The implementation of Six Sigma is helping SAIC measure and improve the operational performance on all of its projects—no matter what their size or complexity. The goal is for all stakeholders at every level of the business to be involved in designing and implementing a plan to eliminate process defects and variations. The result of this concerted, aggressive effort is enhanced productivity and cost savings.

Fundamentals of Six Sigma

The principles of Six Sigma can be applied to simplify the work order process for virtually any type of project.

The Six Sigma approach identifies and eliminates defects with a structured, data-driven, problem-solving method of using rigorous data gathering and statistical analysis. Six Sigma differs from traditional quality improvement programs in its focus on input variables and root-cause analyses, thereby eliminating the need for unnecessary inspection and rework processes.

The three key characteristics of this approach are managing decisions with data, providing training and cultural change, and securing a firm commitment from the leadership of the organization.

"Sustained success requires a commitment from the leadership and it must be practiced by everyone," says Tedaldi. "Improved performance is inextricably linked to an investment in professional-level training. New ways of thinking, communicating, and operating must become part of an organization's culture."

The first step in applying Six Sigma is to define the problem that needs to be solved. A project team is established to ensure the focus is not overly broad and is defined in a business case that establishes the objectives of the project. Preliminary process maps are developed at this stage.

"If you're not making process maps, then you're not doing Six Sigma," adds Tedaldi. "Process mapping enables you to locate bottlenecks, determine how much time it takes to do each step, and measure all of these different steps."

The next step in Six Sigma is the process benchmarking stage where cycle times are identified and performance data is gathered. The analsis phase outlines when and where problems, defects, and inefficiencies occur and defines performance objectives.

The final stages are determining how process capabilities can be improved and identifying controls that can be put into place to effect change and sustain the improvements.

"An implementation and monitoring plan with behavioral analysis to ensure acceptance by users is critical to success," notes Tedaldi. "The plan will include documented procedures, metric targets, a response plan for management, and transfer of responsibility for outcomes. The goal of these steps in the Six Sigma process is achieving sustained, improved output."

Putting Six Sigma into Action

An up-to-date process map was prepared by SAIC and it became clear that the procedure defining work order closeout was inadequate. More than 90 steps and 10 decision makers were required between the work order request and the final closeout. A revised closeout procedure is now used and a new, single-page closeout data report satisfies the information needs of both NCI and SAIC.

Cycle times on small capitalization projects valued at less than $50,000 have improved from about 100 days to 60 days. Initial estimates of how much a project will cost are also improving with most jobs being completed right on the money.

"Funding is really limited in the government, so it is important to meet your targets," says Tedaldi. "In the past, we tended to overestimate jobs by up to 20 percent. Now, we are plus or minus five percent on our initial estimates at 100 percent design."

Most subcontract estimates are coming in at plus or minus 15 percent, with a considerable portion within about five percent of the awarded value. The new Six Sigma processes reduce variability and increase predictability of the SAIC operations. Schedule performance is important, also.

"When we have researchers coming into a lab, we need to get them in on time," says Tedaldi. "We are meeting our current schedules. It is very important for us to meet that target. Over the past 12 months, more than 97 percent of all projects were completed on or ahead of schedule."

An examination of primary source data quality revealed that a data dictionary was needed to ensure compatibility of definitions and reproducibility of results. Establishment of clear definitions of basic conditions enables the customer and contractor to set expectations for performance measurement. Numeric targets help estimate accuracy of the initial estimates versus costs at completion, and the relative contributions of engineering, project management, and subcontractor oversight versus total project cost.

"Analysis of system data indicated that manual entry problems had compromised some data sets due to missing or incorrect entries," explains Tedaldi. "Increased automation, consolidation of data management systems, and training have reduced these errors and provide for an accurate and useful data set. To improve customer satisfaction and assign accountability for changes to scope, the concept of incorporating a project team has been introduced. This fully inclusive team approach has been an extremely effective means to reach consensus on project scope early on and, thereby, eliminate costly changes downstream."

Customer surveys, submitted at the end of each work order and summarized at the end of each award-fee period, are used to focus the team on performance issues most important to customers. The improvements have been measurable with a doubling of award fees since the inception of the Six Sigma effort two years ago.

Preventive maintenance crews are working to improve coordination with lab users. This has increased their ability to complete more than 98 percent of all projects within the assigned period in spite of restrictions on access to the facilities. SAIC crews are meeting their performance schedules and customer satisfaction is increasing.

Working as a Team

Project teams are critical to the success of an organization and must include all stakeholders. Credit for a successful project should accrue to the entire team and not individual members.

"Corporate skills, such as meeting deadlines, cost accounting, and multitasking, should not be exclusive to the business world, but often receive a subordinate ranking in the hierarchy of academic priorities," says Tedaldi. "In academia, individuals compete for credit and although lab groups may pursue common goals, they do not operate as true teams."

Businesses typically approach projects by focusing on relevant facts and networking rather than working from the ground up.

"Understanding the fundamental differences between the two and accepting these as valid within the context of their respective fields can help discussions move forward," adds Tedaldi. "Academicians must be willing to take time from their research to ensure that their expectations are communicated through industry-accepted vehicles such as project scoping documents and work plans. They must realize that in-process change—no matter how necessary to the intended research—inevitably carries a cost in schedule and dollars."

Success and Challenges

SAIC achieved its biggest productivity gains by using a team approach and concentrating on process improvements in the areas of alternative acquisition strategies, enhanced progress reporting and communications, revised estimating procedures, work order prioritization, central management of databases, and integration of safety, procurement, and facilities.

"We have proof that our activities and our management initiatives work. It's not just talk," says Tedaldi.

Looking to the future, the desire for industry-wide benchmarks is high, but comparisons are difficult due to variances in data collection, coding, and analysis.

"We may all have the same measurement techniques, but maybe another company's data aren't that good," says Tedaldi. "Standards for metrics should be established by an industry council."

By Tracy Carbasho