This article reviews that designers and managers are taking aircraft maintainability more seriously than ever. For the first time, the concerns of the maintainers are being addressed early in the design phase, before any hardware is produced. Maintainability is one of the key criteria by which the US Department of Defense will award the initial production contract. Given the new emphasis on maintenance, engineers are investing extra effort to predict that key tasks can be done efficiently. Lockheed Martin has been able to back-fit to the F-16 and F-22 programs some of what it learned from the JSF simulations and animations. Both those planes were designed before maintainability became such a target. Simulations are regularly shared with key partners, such as those involved in JSF propulsion. The first JSF simulation took nine months—a combination of learning curve and time needed to model thousands of postures and movements by four human models.
Units of Lockheed martin and Boeing, in a two-horse race to be the supplier of America’s next-generation combat aircraft, the Joint Strike Fighter, have had to add a new consideration to the issues that will guide their design.
Designers and managers are taking aircraft maintainability more seriously than ever. And, for the first time, the concerns of the maintainers are being addressed early in the design phase, before any hardware is produced.
Maintainability is one of the key criteria by which the U.S. Department of Defense will award the initial production contract. At the heart of the efforts in both companies are advisory panels of senior noncommissioned officers from all the services, along with painstakingly detailed computer simulations, in an aircraft design initiative called “maintainer in the loop.”
Considering the maintainers’ needs isn’t new. But with earlier fighters their concerns were addressed relatively late in the design process. As a result, many maintenance problems were built into the aircraft.
Weapons bays and sections of the planes often accessed for repair were recreated in sheet metal, plywood, or cardboard. The mockups were rough approximations of the real thing, and many didn’t reflect the latest design changes.
Computer simulations were of limited usefulness. Lacking access to design files, many simulations were manually built from best-available dimensions. Results were reported in charts and graphs. The process was slow. By the time the results finally came in, the smallest changes could cost tens of thousands of dollars.
Given the new emphasis on maintenance, engineers are investing extra effort to predict that key tasks can be done efficiently. For instance, Mike Barron of Lockheed Martin Aeronautics Co. in Fort Worth, Texas, has simulated removal of the Pratt & Whitney engine from the JSF demonstrator to verify to the U.S. Navy that maintainers working in cramped spaces aboard aircraft carriers can do it.
The simulation had a virtual maintenance crew of four plus two helpers to push the engine free of the airframe. This was the first major maintainer-in-the-loop simulation. It helped persuade the Navy to relax a long-standing hangar deck requirement that engines must be removed “within the shadow of the aircraft,” that is, from below.
The engine-removal simulation was used in briefing the crew that installed the engine in the concept demonstrator aircraft, the X-35. This first installation was done in three hours, and the crew credited the simulation for part of their preparation.
The simulation of an integrated combat turn, the process of inspecting a plane between sorties, includes a weapons bay and shows the complex interactions of 11 ground crew members and the pilot conducting the necessary tasks between landing and takeoff. The simulation focused on loading Mauser BK-27 cannon ammunition and ensuring access to the aircraft’s maintenance interface panel.
A simulation in Delmia’s Envision software was used to program the JSF gantry automated drilling system, a huge, specially designed vertical fixture for drilling and milling the wing carry-through section. Because the wing carry-through is the aircraft’s biggest and most critical structural component, the simulation showed an entire section of the factory, including the material handling that would be required.
This article was prepared by staff writers in collaboration with outside contributors.
The simulation shortened work time. What was to have been three days’ work by four workers was done in 30 minutes by three. Handling components like this, the so-called “big-bone” parts, is critical if Lockheed Martin is to meet its initial goal of producing the JSF in five months (compared with 12 to 24 months for current fighters).
“We are helping to define the maintainability aspects of the aircraft and its support on the ground or aboard ship,” explained Barron, whose title is specialist, JSF maintainability for human factors. “We are trying to influence the design of the plane—early on, nose to tail—in the interests of the maintainer. “The three big drivers in our work are ensuring physical access with human joint-and-body movement, ensuring visual access, and ensuring access for the necessary tools,” he said.
In modeling commonly used tools, Lockheed Martin follows the Snap-on Tools Corp. catalog.
Federal standards define both the maintenance tasks and the people performing them, with the human models characterized in terms of physical body percentiles.
Prior to the JSF program, Lockheed Martin strove to make maintainers’ tasks doable for a huge range of human physiques, from a 5th percentile female (smaller than 95 percent of all women) to a 95th percentile male (larger than all but 5 percent of men).
“However, using this percentile data did not always guarantee the accessibility sought by the customer,” Barron said. “So the government changed the requirement to accommodating four anthropometric cases, one small person, one large person, and two short, stocky persons. This more realistically represents 90 percent of the maintainer population, and all future JSF simulations will use these new cases.” Michael T. Golas, Barron’s co-worker and manager of Lockheed Martin’s maintainer-in-the-loop program, said, “The simulations are affecting the way the designers go about their tasks. The designers now solicit our advice. It was not unusual to see weapons people or structures designers lined up at Mike’s desk when he came in to work.” Mark Boudah, a career maintainer now working on reliability and maintenance simulations from the famed Lockheed Martin Skunk Works in Palmdale, Calif., said paper and pencil were the usual calculation method in the past.
“Until we got design data integrated into simulation for maintainability, we used milspecs for guidance, added an appropriate percentile male or female maintained and did the work on paper,” Boudah recalled. He is JSF’s liaison with the group of noncommissioned officers who advise Lockheed Martin on maintainability. The true experts, they work though a Maintainability Systems Advisory Panel.
Lockheed Martin has a major campaign to upgrade engineering technology and methods of design (conceptual, preliminary, and detailed) called the Virtual Product Development Initiative. Its mandate extends through the documentation for assembly, known as the build-to-print package, and even to field operations.
Lockheed Martin has been able to back-fit to the F-16 and F-22 programs some of what it learned from the JSF simulations and animations. Both those planes were designed before maintainability became such a target.
Simulations are regularly shared with key partners, such as those involved in JSF propulsion. According to Dana Phelps, who is responsible for the reliability and maintenance factors in propulsion integration—that is, connecting the engines to the airframe—“We paste screen snapshots into PowerPoint documents and shoot them off to Rolls-Royce in the U.K. They send them back with the modifications made and changes noted.”
Rolls-Royce teamed with Pratt & Whitney Aircraft’s military engine works in West Palm Beach, Fla., to provide the shaft-driven, lift-fan propulsion system for the short-takeoff, vertical-landing variant of the Lockheed Martin JSF. General Electric Co.’s Aircraft Engine Group will produce an alternate version of the JSF engine in Lynn, Mass.
If there is a frustration with JSF simulations at Lockheed Martin, it is the seemingly limitless flexibility of human limbs and torsos and the time needed to fully represent maintainers ergonomically and anthropometrically. A typical maintenance simulation can take up to four months because even simple tasks require that hundreds of individual postures be modeled.
In turn, a posture may be composed of dozens of individual step points, or angle changes, in individual body parts—from fingers to torsos. Many full-body activities, such as climbing, twisting, bending, or reaching, require step points for dozens of body parts. The work becomes especially complex and demanding when two or more human models must interact with each other.
The first JSF simulation took nine months—a combination of learning curve and time needed to model thousands of postures and movements by four human models. Despite having three times as many human models, the second simulation was done in six months.
Barron’s simulations run under Unix on an Octane workstation from Silicon Graphics Inc. of Mountain View, Calif., with dual 250-MHz CPUs, 512 megabytes of RAM, and two 9-gigabyte disk drives.
Perhaps Mark Boudah summed up simulation’s benefits best.
“With the ability to perform dynamic simulations, we now can use the designers’ own creations, integrated with other parts of the design, to show them what our issues are. We use their electrons and their data, which is most persuasive. Sure, we did a good job of maintainability on previous aircraft,” Boudah continued, “but it was all done after the fact, after the metal had been cut and riveted together.”