Computers make our modern lifestyle possible, processing mountains of data at the speed of light. However, computers rely on something known as binary processing, meaning something either “is” or “is not.”
Complex software code is required to address all possible environments, factors, and considerations in order to develop a solution. Conversely, humans can come up with very good solutions using incomplete data and inexact methods.
For example, when you have an important appointment, you estimate the time it will take to travel to your destination to avoid being late. Without having exact data, you approximate and combine various factors such as distance, weather, traffic, and other intangibles and derive a “fuzzy inference” that allows you to decide when to depart and what route to take. More often than not, your “guesstimate” gets you to your destination on time.
A similar notion can be applied to the weather. If it is pouring outside, it’s clearly “raining”. If it is sunny and clear, it’s clearly “not raining”.
But, what if it’s sunny and there is an occasional intermittent sprinkle? Is it raining or not? The determination is “fuzzy”. The process of using approximate data and analysis to produce answers encompasses the science of fuzzy logic.
In recent years, a whole new field was born using computational methods that resemble the fuzzy human decision process. In fact, fuzzy logic is at work in many of your consumer goods. For instance, camcorders are now “smart” enough to compensate for operator shaking and movement; televisions have automatic color, brightness and acoustic control based on environmental conditions; and modern automobiles measure conditions such as engine speed and distance traveled to let us know when it’s time to change our oil — compared to the linear model of changing it every 3,000 miles.
Fuzzy logic’s reasoning works by using approximate values rather than exact values to derive answers and solve problems. This methodology is extremely useful and important because so many problems we face are complex and contain incomplete, anecdotal or qualitative information.
Facing complex problems like these are no stranger to the Department of Defense (DoD). In fact, the Weapon System Technology Information Analysis Center (WSTIAC) often leverages fuzzy logic to help the DoD improve their decision making process. For example, WSTIAC employs fuzzy logic and an overarching Business Case Analysis (BCA) framework to address a number of challenges related to weapon system technology. The BCA considers quantifiable and non-quantifiable factors to assess alternative courses of action (COA) and derive the optimum COA in order to solve the problem. In the absence of a complete data set, WSTIAC employs fuzzy logic by leveraging a panel of Subject Matter Experts (also referred to as “using Delphi Methods”) and linguistics, structured, and rule-driven Decision Support Maps (DSM). All this is done in conjunction with multi-attribute decision theory to derive optimal outcomes. As shown in the figure below, this method can be employed to address complex problems to high certainty resolution levels using comparatively simple analytical methods.
WSTIAC has effectively employed these methods to address problems facing the DoD community in areas as diverse as determining the viability of the ammunition industrial base, assessing payoffs from individual weapon systems, evaluating alternative weapons development strategies, assessing defense acquisition alternatives, and others.
The Weapon System Technology Information Analysis Center (WSTIAC) is one of ten Information Analysis Centers (IACs) established by DOD and managed by the Defense Technical Information Center. WSTIAC is a Center of Excellence responsible for acquiring, archiving, analyzing, synthesizing, and disseminating scientific and technical information related to conventional and directed energy weapon systems, their development, production, fielding and maintenance.
Interested in learning more or working with WSTIAC on an upcoming effort? We can be reached via the IAC website http://iac.dtic.mil