At the Museum of Science Boston, visitors discover the wonders of the human body by immersing themselves in a series of interactive activities in which they use their own bodies as the test subject. The Hall of Human Life was uniquely constructed by our in-house designers with the process of learning at its core. A wristband marked with a unique barcode allows visitors to take measurements of their own body, record their experiences, and see how their data compares to other visitors. Thousands of data points are generated each day from the Hall of Human Life. In turn, we can use the data to constantly improve the exhibition and ensure each station is functioning correctly.
From the onset, we knew that technology would be the keystone of the Hall of Human Life and that it would require updating as technological capabilities evolved. When the exhibit opened in 2013, we were hosting the data on a Microsoft SQL Server database with a basic Excel dashboard for reporting and analytics. This on premise system allowed for the capture and retention of all inputted data, but was limited to sampling from the last 30 days of data for display in the exhibit. As of October 2016, anonymous data for nearly one million visitors had been recorded. However, because not all of it was accessible in real-time, the entire dataset was not being used in the truly comparative way that was central to the exhibit’s mission.
Like any public institution, we host large groups of visitors every day. This includes family groups, class field trips, Girl Scout troops, Boy Scout Councils, summer camps, or even overnights for elementary and middle school students. Imagine a large group of elementary school students exploring the exhibit and entering their data into the kiosks. When the next visitors come in, they will not be offered an accurate spread of human data. The comparison will be reflective only of the large group of students that previously went through. This was a real problem because the Hall of Human Life was created to harness the true power of comparative data for interactive learning.
To address this technology challenge, we worked with our friends at Microsoft. Through thoughtful collaboration with our staff, they assessed our needs and helped us to complete the Hall of Human Life in a way that was most useful to us for the long-term. Under the direction of Microsoft engineers, a team of interns from Worcester Polytechnic Institute built a prototype system that allows us to host all the Hall of Human Life data securely in the Azure cloud and recall it all back to visitors in real-time. They also designed a Power BI operations dashboard that monitors the exhibit in real-time and an anomaly detection system in Azure Machine Learning that automatically detects hardware failures and outlier data to ensure the data we are collecting is accurate.
Through our partnership, we have fully updated and modernized the visitor-facing technology in the Hall of Human Life. Now this interactive exhibition not only demonstrates the power of immersive learning anchored in comparative data but it is also a model for transformational corporate-civic partnerships.
Tags: azure, Azure Machine Learning, Boy Scout, Girl Scout, Hall of Human Life, machine learning, microsoft, Microsoft Azure, Microsoft New England, Microsoft Power BI, Microsoft SQL, Museum of Science, Museum of Science Boston, New England, Power BI, SQL, Worcester Polytechnic Institute