Predicting Natural Disasters and Other Human Disturbances
The Los Angeles Police Department is using predictive policing software to determine when and where crime is most likely to occur next. Behavioral information is used to target physical locations, cutting the city into 500' by 500' boxes "Policing in a Box." The technology uses earthquake prediction algorithms to generate a set of results. Do natural and human forms of destruction really follow the same pattern?
Predicting Natural Disasters and Other Human Disturbances is a project that examines the relationship between information and form. Using the predictive policing algorithm as inspiration I collected two data sets: Los Angeles earthquakes and burglaries in a specific neighborhood over a two week period. How can this data set be used to express the correlation beyond data visualization. How could it still communicate but express something more? Could the information be used to generate speculative forms, narratives, buildings, or city plans?
Poster
Using Processing to sort the two data sets I generated an image [photo above] that showed the correlated pattern of natural and human disturbances. The two weeks of earthquake is on top and burglary data is on the bottom. I was surprised that the two seemingly different data sets generated a very similar pattern.
I learned that after an earthquake, many aftershocks follow. This image revealed to me that burglaries are quite similar: A large set of robberies will occur in a neighborhood, mimicking an earthquake, and then they become smaller and less frequent, like aftershocks.
Speculative Urban Model
How could this information be used? Looking at work by BIG, our next steps were to take the two-dimensional processing image and translate it into three-dimensional form.
What will the next city plans for Los Angeles be designed utilizing this data? This speculative model theorizes that city planners will utilize predictive data to change the size and spacing of high crime and low crime city blocks. City blocks with high predicted crime might be narrowed, making it easier for police to patrol. Low crime blocks might gain extra space, forming petrographical barriers.
I was really fascinated by this process of using data to explore more abstracted form, something inspired by rather than just a data visualization. My background is in industrial design and I got really excited about utilizing this process to generate forms at a different scale.
Process for Poster
My first Processing sketch using earthquake and burglary data hinted to a similarity however the spacing between each data point did not help emphasize the point I was hoping to illustrated. I decided to increase increase the spacing between each point and create larger triangles that would have more visual impact at a glance while being a bit more abstract.
Process for Model
I used the triangle shapes generated in processing to create the city blocks. Exporting the pattern from illustrator into rhino, I extruded the forms to help myself visualize the three-dimensional forms and angles. I printed the forms from rhino and tapes onto wood to create a template to cut on the ban saw.
At first I thought I would CNC the forms, creating one fully connect piece. However there wasn't a lot of time. The decision to make individual pieces for each payed off in the end. They were easy to move and play around with the spacing between each. I like how they referenced child's play blocks and hinted to the city planners process, like planning a war, each block would be moved closer or further away.
Lab Core A: Structures
Instructor: Timothy Durfee
Processing Workshops: Casey Anderson
Processing and Rhino were used to visualize and form data.
Art Center College of Design / MDP