Define your scope or domain where the use case is relevant or prevalent?
Road condition in Michigan
What is your main story?
This project has two folds: 1) the big picture: the project proposes to link the road conditions rated by the PASER system to the location (zipcode) and social-economic data (e.g. income, crime rate, race data from Census). The outcome may discover a correlation between the road conditions and the location and the social-economic status of residents.
2) the user-generated case/ individual story: users of the case could enter the data (text and visual) of the road conditions of their neighborhood and commute routes. By finding out the road condition rating (via questionnaire) and how much they pay to fix the car and their personal data, the project could identify the patterns and the consequences of road conditions through individual stories.
Self-reported data:
- Do you have to swerve in order to miss a potential tire blowout?
- Are your raods filled with black tar lines for temporary repairs?
- Is your daily commute typically bumpy/ require you to drive slower?
- Has your commute ever led you to make repairs to your vehicle?
- Is your commute filled with construction zones and potholes?
- SCORES: 0-4: Good/Fair 5-7: Fair/Poor 8-10: Poor
Who are the characters or people in the main story?
a regular commuter: could be a student, faculty, or a resident who commute on a daily basis.
What happens?
The roads in Michigan are notoriously bad, which has caused vehicle damage, expensive repairs, and safety concerns. The governor has proposed $.45 per gallon gas tax to fund the road repair but didn’t pass the bill. While the government seeks alternatives, the commuters have to deal with the horrible road conditions and the consequences. The commuters in just 8 areas in Michigan spent an average of $500-$800 per driver for vehicle wear and tear due to bad road conditions. The road conditions in some neighborhoods and areas may be worse than the others. See the map here: https://maps.semcog.org/PavementCondition/.
Where?
the Metro Detroit area in Michigan, and then the larger Michigan.
When?
1) project 1: depends on how far the data go back. https://semcog.org/pavement
2) project 2: the current student’s and residents’ experience in the most recent year.
Why?
The bad road conditions affect every commuter, but the consequence may have a differential impact on individuals with different social-economic status. The project is meant to identify the pattern between the road conditions and the wellness of the neighborhood/commuters. The road conditions also have been on the top agenda on the government’s budget for years and yet left unsettled due to disagreement on where money should come from and by how much. The community has invested interest in the topic and the conversation should continue to brainstorm the ways to fix the roads.
How?
See the section on the main story
Exercise if you can break a big story to several smaller stories
- Where (zipcode/town/city) has the worst road conditions according to the PASER rating?
- Who lives in those areas (income, education level, race, crime rate)?
- Is there any correlation between road conditions and the neighborhood?
- Is there any significant difference in the road conditions among the neighborhoods?
- How much do the commuters spend on fixing their vehicles due to bad road conditions?
Exercise if you can put the smaller stories into a sequence of stories, or in other words, a sequence of decision-making process.
- Step 1: where (zip code/town/city) has the worst road conditions according to the PASER rating?
- Step 2: Who lives in those areas (income, education level, race, crime rate)?
- Step 3: Is there any correlation between road conditions and the neighborhood?
- Step 4: Is there any significant difference in the road conditions among the neighborhoods?
- Step 5: How much do the commuters spend on fixing their vehicles due to bad road conditions?
Exercise if you can extract the important information (age, ethnicity, grade, preference, etc.) from 5W 1H of your main stories.
- zipcode
- road conditions rating
- the median income of the zip code/city
- the crime rate of the neighborhood
- the racial ratio of the city
- the expense of fixing the vehicle due to bad road conditions
Explain what you expect from ML applications for each small story (for example, prediction, recommendation, analysis, visualization, search, topic, modeling).
- I expect the ML to search the road conditions according to the zip code.
- I expect the ML to visualize the road conditions according to the zip code.
- I expect the ML can correlate the crime rate with the road rating in each zip code.
- I expect the ML can correlate the racial ratio of the city to the road rating.
- I expect the ML can correlate the median income level of the city to the average road rating.
- I expect the ML can calculate the estimated expense of fixing the car based on the road condition of the area.
Story By:
You Li
