Cost estimation is an art. Overestimate and you can lose the job you are pitching. Underestimate and you may suffer financially. Such is the complexity that a whole raft of specialist companies have popped up over the last 25 years to provide tools and expertise, not just for small and medium sized businesses, but for the largest corporations as well. Galorath (El Segundo, California; www.galorath.com), is one of these providers, having earned its laurels in the highly competitive and cut-throat defense and aerospace industries. Its customers include Agusta Westland, Airbus, BAE Systems, Boeing, Fokker, GKN Aerospace, Kongsberg, Lockheed Martin, Northrop Grumman and the U.S. Navy.
“The company was started in the 1980s by Dan Galorath to develop cost-estimating software,” says Carl Dalton, the company’s UK-based managing director, International Operations. “Dan was a software project manager who, on seeing that there was a need for it, developed SEER-SEM: System Evaluation and Estimation of Resources—Software Estimation Model.” SEER-SEM is a powerful decision-support tool for estimating software development and maintenance cost, labor, staffing, schedule, reliability and risk as a function of size, technology, and any project management constraints. It is effective for all types of software projects, from commercial IT business applications to real-time embedded systems.
“SEER-SEM gives you the power to change a project before its budget gets out of control,” says Dalton, “because it can identify cost drivers such as staff and schedule constraints, specification volatility, the impact of using new versus re-used software, and development methods and standards.” Over the years, the SEER product line has expanded to include a software sizing model (SEER-SSM), integrated circuit model (SEER-IC), hardware estimation model (SEER-H) and hardware life cycle model (SEER-HLC). Says Dalton of industry acceptance of this suite of tools: “They are widely used within defense, military/aerospace, government and the banking, finance and insurance industries.”
Unique to the SEER suite, however, are their knowledge bases (Kbases). “They are created from real, completed estimates that are detailed, quantified and repeatable, providing users with an instant baseline to define, refine and measure against,” says Dalton. More than 100 Kbases are available from which to create thousands of possible scenarios. This lets the client pick one that is appropriate for the system he wants to develop and the standards to which he is working. He also can create a knowledge base from historical data to further fine-tune the estimating process.
“For example,” says Dalton, “it questions whether it is an existing model that needs modification, and how it can be re-used.” From there the model calculates reusability and rework, looks at the people involved and their experience level, evaluates different environments and how robust the requirements are, and considers whether or not the result has to interface with hardware and the target platform. It also judges whether and how much testing is required, and gives an indication of the maintenance schedule. “It may take quite a long to time produce the first estimate because the client is setting the model up, but once done,” says Dalton, “he can re-use the pre-defined template, altering different elements such as the size, the material or alternative processes as necessary. One of the benefits is that the template can be copied to produce another estimate, creating up to six comparisons.”
This means the different options can include sensitivity analysis, risk analysis, as well as trade-offs in the different manufacturing processes. “You can even consider having something built in another country which possibly has lower manufacturing rates,” says Dalton. Thus it’s possible to consider capability levels versus rates in order to assess quality and cost, and inspections and other non-core items can be added to the estimate equation. The software’s other strength is its ability to accept information from outside the system, such as Excel spreadsheets. “We’ve had some success linking it to CAD systems such as SolidWorks, and it has also been linked into Catia V5 on the composites side,” says Dalton. “It can also link into Middleware for stress calculations and into Engineous Software’s iSIGHT-FD that helps integrate and automate leading CAD and CAE tools. As you make changes you can actually see the effect.”
ATK Composite Structures (Edina, MN; www.atk.com) is a major composites subcontractor—its products include composite structures for the Joint Strike Fighter, rocket motors, and F-22 horizontal stabilators—that prepares 80-90 proposals per year. Having previously relied on a cumbersome, bottoms-up estimating approach that required spreadsheets with hundreds of lines and a number of skilled estimators—the quality of whom varied—to work it all out, ATK turned to SEER-DFM with Composites Plug-in, a tool that emerged in response to the aerospace industry’s Composites Affordability Initiative (CAI). ATK chose it specifically for the composites module, which offers 25 existing and emerging composite fabrication and assembly processes, as well as some state-of-the-art metal processes. The composites-specific tool enables engineers and analysts to understand the impact of critical cost elements on design features and process parameters specific to the composites environment. As a result, designers working on new concepts can see the cost impact of their decisions in minutes.
Since bringing in SEER-DFM nearly four years ago, ATK has completed about 240 proposals. During that time, the tool has allowed the company to streamline its proposal process significantly, saving around two to four days per proposal. “The end result for ATK,” says Dalton, “is that it reduced the number of employees it included on proposals from about nine people to four.” With no less than 80 proposals a year, moving to Galorath’s software has saved nearly 300 days annually, and eliminated the constant back-and-forth between the project team and management to achieve consensus that lead to inconsistency between proposals.