Overview
JHB Municipal Infrastructure Intelligence is a portfolio-grade civic analytics project simulating how a City of Johannesburg infrastructure team could track and prioritize service delivery using evidence-based KPIs and a Python-driven analysis pipeline.
Intelligence Layer
Johannesburg faces recurring infrastructure pressure: aging road surfaces, burst water mains, and reactive-only maintenance cycles. Without an analytics layer, ward managers and operations leads cannot distinguish systemic failure zones from isolated incidents — and cannot hold contractors accountable to SLA expectations.
Problem
Municipal infrastructure data — potholes, burst pipes, water leaks — is fragmented across departments, making it difficult to prioritize repairs, monitor contractor SLA compliance, or identify repeat-incident zones before they become systemic failures.
Data / Signals
Analyst Objective
Design a Python analytics pipeline that enables infrastructure teams to:
- monitor SLA compliance by incident type, ward, and contractor,
- detect repeat-incident hotspots before they become systemic,
- and track backlog trends over time for prioritization.
Stakeholders
- Ward managers needing operational visibility by area.
- Infrastructure operations leads monitoring contractor performance.
- Civic oversight teams requiring SLA compliance evidence.
Key Questions
- Which wards have the highest SLA breach rates and why?
- Where are pothole and water leak incidents clustering into repeat-problem zones?
- What is the relationship between response time and resolution quality?
- How does backlog trend over time — and which contractor is contributing most to it?
KPI Framework
- Response: median time-to-first-response (hours) by incident type and ward.
- Resolution: mean time-to-repair (days), SLA compliance rate, breach rate by contractor.
- Backlog: open incident count trend, repeat-incident rate per street and pipe zone.
- Hotspots: top 10 wards, suburbs, and streets by incident density and SLA breach frequency.
Insight
- A relational synthetic dataset (200,000 rows) was structured to mirror real JHB operational patterns — messy, realistic, and BI-ready.
- A modular Python pipeline covers data generation, cleaning, EDA, KPI calculation, and hotspot detection across all pipeline stages.
- Analysis output is structured as an interactive HTML dashboard for immediate stakeholder readability.
- Every KPI output is framed around a specific operational decision — not just a description.
Implication
- SLA breach rates varied significantly by contractor and ward, pointing to both resourcing gaps and accountability failures.
- Repeat-incident patterns in high-density wards indicated infrastructure asset deterioration, not random failure.
- Backlog trend analysis revealed that reactive-only maintenance creates compounding service debt over time.
Closing
Deliverables
- Python pipeline covering data generation, cleaning, EDA, and KPI output.
- Interactive HTML dashboard for operational and stakeholder review.
- Ward and suburb-level hotspot detection layer.
- KPI framework aligned to real municipal service delivery standards.
Outcome
The project demonstrates how analyst-grade Python work translates into decision-ready civic intelligence — moving from raw incident logs to prioritized, SLA-aware, ward-level operational insight.
Link
- Live dashboard: View live dashboard
- GitHub repo: Open GitHub project