academic analytic competition presentation
TRANSCRIPT
Advanced Analytics Competition
Martin SellersLintsen Han
Scarlett ChengJeffery RyanVince Rich
Executive Summary andResearch Question
Research Problem: Scarcity of Public Resources for City Building Permitting and Violations
Goal: To create a operational model to identify which buildings are most likely to have violations upon inspection
Model Input Variables:
• Permit Type Risk Rating
• Geospatial Significance
• Permit Issue Date Cycle Time
Rating Model
Permit Type Geography Time of
IssuePriority Rating
Permit Types Risk Rating
• Each permit type was given a priority for inspection of High, Medium or Low.
• Priority was determined by correlating inspection failures with issued permit types.
• High Priority: at least 5% more failures than average• Medium Priority: Within 5% of the overall percent failed• Low Priority: at least 5% less failures than average
Kernal Density Analysis of Building
Violations
Graphical representation of building violations within Chicago utilizing the density study to illustrate the magnitude of these spatial relationships• The findings of this density analysis is consistent with
the findings of Getis-Ord Gi*
• Wards Most at Risk• 28• 24• 17• 48• 42
• Wards Least at Risk• 1• 27• 47
Getis-Ord Gi* Hot-Spot Analysis of
Building ViolationsThe measure identifies the statistical significance of spatial clusters. A cluster with a high value can be interesting but may not be a statistically significant hot spot. To be a statistically significant hot spot, a cluster will have a high value and be surrounded by other features with high values as well.
• Several distinct geospatial clusters indicating violation hot spots
• Priority should be placed geographically on hot spots, these areas are indicative of more violations occurring
• Cold spots indicate areas where low priority should be placed as a result of a distinct lack of significance
• Areas of Responsibility should be standardized to align with other city functions, however more analysis is needed to determine geographic constraints incorporating resource allocation optimization, e.g. human resources
Permit Issued Cycle Time
Time period% of the amount of issued
permitsCumulative Percentage
0-3 Years 20.9% 20.9%
3-7 Years 30.7% 51.6%
7-10 Years 35.5% 87.1%
10+ Years 12.9% 100%
Practical Model Application
Core Measures Operational Output
Permit ID Permit_Indicator Geospatial_Signfigance Cycle_Time Prioritization_Score
1 1 1 2 0.402 3 2 2 0.703 2 2 4 0.804 1 3 1 0.50
5 3 1 1 0.50
Measure Key Index Score
1 = High 1 = High Signfigance 1 = 0 - 3 yrs 0 - .3 = High
2= Medium 2 = No Significagance 2 = 3 -7 yrs .3 - .7 = Medium
3 = Low 3 = Negative Signficgance 3 = 7 - 10 yrs .7 - 1 = Low
4 = 10+ yrs
Policy Implications For the City of Chicago
• More Efficient Resource Allocation
• Increased Public Safety
• Increased Community Awareness
• Scalability