Support WU
A-Z Index
Past Projects



Atkinson Graduate School of Management

Willamette University
900 State Street
Salem, Oregon 97301

503-370-6167 voice

Design Thinking

Inter-driver communication

There is a limited language of communication that allows us to convey our intentions while driving. Things like “I am turning right”, or “Warning, I am slowing down” can be communicated with turn signals and brake lights. The problem is that beyond the bare essentials, there are few options available to drivers to communicate non-essential messages like “I am late, please let me pass”, or “Sorry, I didn’t see you in my blind spot”. It is this lack of non-essential communication that adds to the frustration of driving, and is likely a source of the road rage that is common today. Drivers often resort to hand gestures or even aggressive driving tactics to try to convey a message,  usually resulting in a distorted message being received. The project team researched what messages were most needed, and then designed and tested a series of simple language-independent signs that could be displayed on a LED matrix mounted on the front and back of a car. Early testing found them well-received.

A dirty clothes detector

A surprising number of people (usually children and college students) organize clothing that is somewhere in the continuum between worn-once and needing to be laundered in piles on the floor of their living space. How can they tell when clothes should be laundered before the smell-test makes it obvious? How can an athlete choose clean equipment in the locker room? How can a mother optimize her laundry chore? This project team designed a marker-like pen filled with a chemical that turns color when exposed to certain concentrations of the salts left behind by human sweat. They also saw potential for the product as a water-saving aid in contexts where employees wear uniforms and companies are responsible for laundry.


Data Mining

Forecasting the wind

Renewable energy sources are the next wave of power generation. Currently, many different producers are using windmills to produce energy. Windmills provide an energy source that is sustainable and does not harm the environment. However, it is difficult to predict the wind. Wind energy is also difficult to store, so energy produced by windmills needs to be used as it is produced. Sudden changes in the wind affect an energy company’s ability to maintain stable production levels. Dependable power generation can be assumed when there is a high probability of a wind speed of at least 7 MPH. The project team used weather data to predict future wind speed two hours in the future, six hours in the future, and 24 hours in the future. This level of predictive accuracy gives sufficient lead-time for producers to confidently switch from more costly modes of production to the windfarms.

Interpreting Mammograms

A mammogram is an x-ray picture of the breast. Currently they are the most effective method of breast cancer screening. However, mammograms are difficult to interpret. Annual screening mammograms miss up to 20% of breast cancers present at the time of screening, while 70% of subsequent biopsies completed after an abnormal mammogram are said to be false positive. A false positive result can lead to anxiety and other forms of psychological distress to the patient, nevertheless it is important that healthcare professionals rule out potential cancer. There is a need to better predict malignant breast cancer. The project team analyzed the effects of a wider array of measurements and found that they were able to increase predictive accuracy to nearly 80%.

Predicting the NBA Draft

Late June is always an exciting time for NBA teams as they select from the college players that hope to become professionals. As with any new venture, a business needs to determine if investments are worth the risk they entail. This project developed the player True Value Index (TVI) from various college play statistics to predict the order in which players would be drafted. Their model predicted a players draft spot within an average of nine draft spots 68% of the time. The model is useful to both players and agents. Players who are trying to decide whether to forfeit their amateur status and become professionals can weigh their prospects against those of other players by estimating their draft position. Agents can also use the model to predict the players with the highest career earning potential. There have been many studies that predicted the usefulness of a player based on college statistics, and studies that looked at NCAA championship performance to predict draft position. However, no other study analyzed the full college play statistics to gain the degree of accuracy this one achieved.