“Which streets, pipes, and assets need us first?”
Cities hold decades of inspections, work orders, and resident reports. They are rarely in one place, and rarely used to decide where the next dollar goes. We put them on one map and rank them by risk.
What it looks like
Repair budgets follow complaints and age tables rather than risk, so crews fix what is loudest, not what fails next.
Budget follows the loudest complaint
Capital and maintenance plans lean on complaint volume and asset age, not the likelihood and consequence of failure.
Records in separate silos
Inspections, work orders, 311 requests, and GIS layers live in systems that do not share an asset ID.
Crews planned by habit
Routes and schedules repeat year to year because nobody can see where risk has moved.
If any of that sounds familiar, here is how we take it apart.
How we solve it
GIS asset models that combine inspection history, sensor data, and 311 requests into a ranked, mappable work plan.
Join the records to the map
Reconcile inspection history, work orders, sensor readings, and 311 requests to one asset inventory in your GIS.
Model the risk
Score each asset on likelihood and consequence of failure: condition, age, material, criticality, and who depends on it.
Turn it into a work plan
A ranked, mappable list that planners and crews act on, feeding the work-order system you already use.
Keep it current
Scores refresh as new inspections and requests arrive, so the plan moves when conditions do.
Data governance in this work
Location data is personal data more often than it looks.
- Resident reports de-identified before analysis, with addresses generalized wherever precision is not needed.
- Every risk score explainable: which records drove it, and how they were weighted.
- Access by role, so field, planning, and public views each see what they should.
What you get
- One asset inventory in your GIS, joined across systems
- A risk model with its method documented
- A ranked capital and maintenance plan
- Dashboards for council and public works
Knowing the approach is half of it. Here is how it gets into production.
Phase by phase, with a gate at each one
Each gate is agreed before its phase begins, so nothing moves forward on optimism.
Assess
Choose one asset class, such as water mains, pavement, or trees, and its data sources.
Gate: Asset class and sources agreedPilot
Build the risk model for one district and check it against known failures.
Gate: Model checked against historyDeploy
Citywide ranking, fed into work orders and the capital plan.
Gate: Crews planning from the rankingExtend
Add the next asset class on the same inventory and method.
Gate: Ongoing
How it runs
Built on the GIS you already run, typically ArcGIS, so planners and crews keep their tools.
Tell us which asset keeps you up at night.
Water mains, pavement, trees, or signals: we will show you what your own records say about where to start.
Other problems we solve
- “Our data is everywhere, and nobody trusts the numbers.”State & local government
- “We need senior technical leadership, not a full-time CTO yet.”Startup founders
- “How do we use AI without putting student data at risk?”Educational institutions
- “Is this model good enough to put in front of the public?”AI evaluation
