The "Hoover Dam" model describes how engineers translate reality into the theoretical world, and how these conceptual designs are then translated back into reality. Sometimes, the theory does not account for all the complexities of the real world.
Through the framing of our bridge project, some information was lost in the translation from the real world to the theoretical world, and this caused our bridge to fail in a way that we did not anticipate. We predicted the bridge to fail due to compression at the top of the bridge, but it ended up failing at the splice connection (where two sections of our bridge were joined together), as shown in the photo below.
Photo of the splice connection failure point
Graphs created using our Python code; the red lines represent the capacities (maximum load that our design can withstand) and the blue lines show the load envelopes (maximum forces that parts of the bridge will be subjected to)
This failure mode was one that we had not accounted for in our calculations. Because a large part of the CIV102 course content centered around performing calculations based on the cross-sectional properties of beams, we implicitly framed the opportunity mainly as a task of optimizing the cross-section and diaphragm spacing of the bridge, but not of optimizing the splice connection. The course content did not include the knowledge involved in calculating the load-bearing capacity of the splice connection either. Hence, we put less thought into the splice connection, and ended up designing the splice connections last. This meant less material was available to design the splice connection. Although other theoretical calculations (such as the the shear force capacity and envelope graphs above) were very useful, the performance of our design was limited by the framing of the opportunity, which excluded certain aspects from our theoretical considerations of the bridge's strength. This eventually led to unexpected setbacks back in the real world, when the bridge was finally tested.