A Small Borough Takes a Data-Driven Approach to Road Maintenance
Canonsburg, Pennsylvania is not a major metropolitan area. With roughly 40 miles of borough roads to maintain and a public works team operating under real budget constraints, the pressure to prioritize repairs efficiently is constant. In April, the borough council approved a $10,000 investment in Vialytics — an AI-powered road assessment platform — to bring objectivity and structure to that prioritization process.
The result is one of the cleaner municipal AI deployments we have examined: a defined problem, a proportionate tool, and a measurable output.
The Problem: Subjective Prioritization and Limited Resources
Road maintenance decisions in small municipalities often rely on informal observation. A public works employee notices a crack on a familiar street. A council member hears complaints from constituents. A section of road gets repaired not because the data supports it, but because someone with influence said so.
Borough Manager Jennifer Foster identified this dynamic directly.
The data is so transparent, so objective, it’s not biased or based on any one person just saying ‘I think we should do this one’ because they say so,
she noted. That kind of institutional bias is difficult to eliminate through process alone — but it becomes structurally harder to sustain when a numerical score exists for every road segment.
The borough also had a secondary motivation: grant applications. Both Washington County and the Pennsylvania Department of Transportation require documented evidence of road conditions to support funding requests. Anecdotal assessments do not satisfy that requirement. Standardized PCI scores do.
The Tool: How Vialytics Works
Vialytics operates through a smartphone mounted to the hood of a vehicle. As the vehicle drives, the phone captures road surface images at regular intervals. Those images are processed by an AI system that analyzes surface conditions and assigns a Pavement Condition Index (PCI) score to each segment.
The PCI is not a proprietary metric invented by Vialytics. It is a standardized quantitative measure aligned with Pennsylvania Department of Transportation standards, running from 1 (immediate intervention required) to 100 (flawless condition). Using an established benchmark rather than a custom scoring system is a deliberate and important design choice — it makes the output legible to government agencies and grant reviewers without translation.
The platform delivers two output formats: a color-coded map of all scanned roadways and an Excel spreadsheet containing the same data in tabular form. Both formats serve different audiences within the same workflow — visual for council presentations, structured for grant documentation.
Hardware and Execution
The scanning setup is intentionally minimal. A smartphone, a vehicle mount, and a driver. No specialized equipment, no external contractors, no complex calibration process. Washington & Jefferson College intern Martin Grissieon handled the majority of the scanning work alongside Foster.
Grissieon described the process as pretty easy. The full 40-mile scan required approximately 32 hours of driving time — a figure that underscores both the accessibility of the tool and the genuine scale of the effort involved. Covering every road segment in a borough systematically is not trivial, even when the technology simplifies the data capture.
Analysis Timeline
The scanning phase is complete. The analysis phase, however, is more involved. Full processing and interpretation of the data is expected to extend into late July. This lag between data capture and actionable output is worth noting — Vialytics is not a real-time diagnostic tool. It is a structured assessment platform that requires time to generate a complete picture.
Preliminary Findings
The early numbers are informative. Over 25 of the borough’s roughly 40 miles of roadway scored 60 or above on the PCI scale. The overall borough average sits at 70, which indicates a road network that is functional but contains meaningful variation — some segments performing well, others requiring near-term attention.
An initial list of the 10 road sections most in need of repair has already been generated from the data. A more comprehensive prioritization list is in development and will be presented to the borough council on July 13 by Grissieon.
Transparency as a Governance Asset
Foster’s plan to publish the scan data on the borough’s website — pending a suitable hosting format — reflects an underappreciated dimension of this deployment. When road condition data is publicly accessible, it changes the nature of constituent conversations. Residents can see the score for their street. Council members can reference the same dataset. The basis for decisions becomes shared rather than opaque.
This is not a feature Vialytics built. It is a governance decision the borough made. But the tool enables it.
The Gap Between Score and Decision
A PCI score tells you the condition of a road surface. It does not tell you the cause of deterioration, the subsurface condition, the traffic load, or the cost-effectiveness of different repair approaches. The data narrows the decision space considerably — but it does not replace engineering judgment for complex cases.
Analysis Latency
The multi-month gap between scanning completion and full analysis output means Vialytics is better suited to annual or biannual planning cycles than to rapid-response maintenance decisions. For a borough operating on a structured budget calendar, this is manageable. For emergency prioritization, it is not the right instrument.
Hosting and Public Access
Foster acknowledged uncertainty about how to make the data publicly accessible in a suitable format. This is a practical friction point that other municipalities considering similar deployments should plan for in advance — the data is only as useful as its accessibility.
Cost-Benefit Assessment
At $10,000 for a borough of this scale, Vialytics represents a defensible expenditure. The scanning required no specialized contractors. The output is directly usable for grant applications, which can return multiples of the initial investment if successful. The labor cost was absorbed largely through an intern placement.
The more significant long-term value lies in repeatability. A single scan establishes a baseline. Subsequent scans — conducted annually or after major weather events — allow the borough to track deterioration rates by segment, measure the impact of repairs, and build a longitudinal record of infrastructure condition. That record compounds in value over time.
Final Recommendation
Vialytics is a well-matched tool for small to mid-sized municipalities that need to move from informal road assessment to structured, defensible data — without the budget or staffing for full-scale infrastructure management platforms.
The Canonsburg deployment demonstrates what a clean implementation looks like: clear ownership, realistic scope, a defined use case (grant documentation and prioritization), and a plan to make the output publicly accessible. The $10,000 price point is low enough that the tool pays for itself if it supports even a single successful grant application.
For public works teams still relying on windshield surveys and institutional memory, Vialytics offers something genuinely useful: a common language for road condition that everyone — engineers, council members, residents, and grant reviewers — can read from the same page.
The data does not make the decision. But it makes the decision harder to make badly.
Comments (0) No comments yet
Want to join this discussion? Login or Register.
No comments yet. Be the first to share your thoughts!