Background Information – an enterprise artificial intelligence company based in San Francisco (SF) – has previously done volunteer work with the SF Government Fix It team to solve critical city problems by leveraging data science. Members of the SF Government believe there is a correlation between crime rate and dimly lit areas caused by streetlight outages. Our group has partnered with to understand the nature of the relationship between these two factors and to explore the possible approaches to repair streetlight outages while possibly reducing crime.


Problem Statement

We first analyzed data – gathered from DataSF – to determine the relationship between street light outages and crime rates in San Francisco. To do so, we generated two weighted and scaled crime rates for each light: one while the light was off and one while the light was on. Then, we used advanced statistical analysis and hypothesis tests to further understand the nature of the relationship. We next used a traditional traveling salesman problem to determine the distance and amount of time necessary to repair street lights in each neighborhood. Finally, we implemented a linear program to obtain a list of neighborhoods whose street light repairs should be prioritized, incorporating estimated time required for each repair, crime rates, and the distribution of broken lights. Regardless of whether or not there is a correlation, street light outages are a common issue reported by San Francisco citizens and therefore need to be addressed.
No page found


Ethical and Social Responsibility

As engineers working on this public service project, our work has the potential to impact thousands of people living or visiting San Francisco, CA. It is our responsibility to analyze the most accurate data and produce exceptional results. If our data analysis proves that there is a relationship between crime rate and street light outages, crime rate will be incorporated as a factor in our optimization of street light repair. We ensure our repair solution will be optimal, fair, and economical, and will be neither intrusive to the residents of the city nor overwork the maintenance staff that will repair the street lights.