Measuring Performance in Human Factors Engineering

From Mckwiki

Jump to: navigation, search

One of the axioms in the energy-saving business is that the first step to take is to measure your energy consumption. Only after you know what you’re using can you make wise decisions about reducing that consumption.

The axiom is: If you can’t measure it, you can’t control it.

This reasoning permeates the next generation of ship design tools: Measuring the performance of the ship.

The Navy has pursued several projects which target the development of MoE (measure of effectiveness) tools in ship design. In most cases, the goal of these projects is to increase the communication between the design community and the requirements community, by permit rapid visualization of the effects of decision-making upon ship design, and vice versa.

One example of such a project is the development of visualization dashboards so that the effects of naval architectural decisions can be seen by requirements-setters. Something along the lines of “See, if I make the ship 5 meters smaller then the ability to do Air Ops goes down.” For projects of this type a major thrust of the effort lies in the discussion of how to measure mission effectiveness. And good work is being done in this thorny subject.

But as of right now none of the ship design models include any parameters for measure the performance of the human factors element. Existing mission MOEs are not capable of reflecting the difference between a human-excellent ship and human-horrible ship - because we don’t know how to measure it.

We propose to undertake the development of measures of human factors engineering effectiveness, which can be included in overall mission effectiveness measures.

Phase 1 - Scoping

The proposal team has some specific initial ideas, but the project would actually start with a preparatory phase of literature search and information gathering. The subject is big, so the right way to eat it is one spoonful at a time. Thus Phase 1 is scoping: Who has touched this in the past? What is the right level of granularity- component, system, supersystem? What is the right breakdown - HFE disciplines, SWBS, other? Is it a nav arch approach or an HFE approach?

One example investigation is the idea of building an area/volume based metric: calculate some areas and volumes to see if spaces are too tight. In addition there are some Objective Functions that can be harvested from the Intelligent Ship Arrangements community to put a “score” on the ship's arrangement. There also exist numerical factors that can be used for ship motions, to put a “score” on ride quality.

Unknown, at the time of this writing, is how to put a “score” on the man/machine interaction? How does one put a “score” on the fact that Ship A is closed and air conditioned so personal protective gear is not needed? This is one area where research would be fruitful.

Another topic that needs to be investigated during start-up is whether the MOE would have to be "component" specific, involving MOE formulas for specific components or processes (e.g., bridge design, anthropometrics, maintainability, egress, noise and motion exposure, etc). The applicable component "scores" could be rolled up to provide a composite value.

If this philosophy is adopted, then it in turn introduces a new difficulty in defining appropriate levels of assessment to obtain this info at different stages of design, especially at very early design stages when it would be most beneficial.

Further, while the authors envision this effort as an HFE effort, it is possible that to be done right it would actually need to be an HSI thing since personnel, manpower, safety, habitability would all need to be incorporated. But perhaps that could be a way to "bin" the assessment, where HFE would have many subsections?

The goal of the proposed program would be a set of metrics that naval architects could actually use. Formulas that could be coded into ASSET… Formulas at various levels of detail so that they can be used in concept design when all we know is Length, but at a higher level of detail they can be used in Preliminary Design to trade off a machinery space arrangement. So the final question is: Are ASSET and LEAPs configured in such a manner that this type of info COULD be incorporated in a useful manner?

I return to the Axiom: If we can’t measure it, we can’t get better at it. And we MUST get better at it, because our Sailors deserve it.

Thus Phase 1 would be a part-year effort of lining up the ducks. The end product of Phase 1 is: "So here's the plan..."

That also serves as an off ramp, in case Phase 1 concludes it's impossible...or that it has already been done.

Another part of Phase 1 - that comes naturally out of the "literature survey" aspect of the phase - is the list of participants for Phase 2.

Phase 2 - Execution

The Phase 2 multi-year execution proposal would be the final product of Phase 1.

Personal tools