{"resourceId":"drjtbc-new-hope-bridge-open-road-tolling-ai-classification","versions":[{"version":"external-1938af20ffe934e61f13924bede8d5d0f4f980933c670a337e15dbfd0cfc1046","resource":{"id":"drjtbc-new-hope-bridge-open-road-tolling-ai-classification","title":"TransCore Launches Open-Road Tolling at DRJTBC's New Hope Bridge","organization":"Delaware River Joint Toll Bridge Commission (DRJTBC); TransCore","sector":"Bi-state toll bridge authority","geography":"New Hope, Pennsylvania / Lambertville, New Jersey, United States","publishedAt":"2026-10-29","publicationDate":"2026-10-29","eventDate":null,"sourceName":"FairsOnline (republished TransCore announcement)","sourceLabel":"FairsOnline news, Transportation & Mobility","sourceUrl":"https://www.fairsonline.org/transcore-launches-open-road-tolling-drjtbcs-new-hope-bridge","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["infrastructure","governance-procurement"],"finding":"TransCore says it has launched open-road all-electronic tolling at the New Hope-Lambertville Toll Bridge (U.S. Route 202), the first of DRJTBC's seven legacy toll bridges to convert, replacing cashless tolling through existing plaza infrastructure. The installation includes TransCore's Infinity Digital Lane System and VCATS, its AI-powered vehicle classification solution. The bridge carries slightly more than 10,000 vehicles a day with nearly 94% E-ZPass use, the highest in the system. TransCore says the change improves safety and reduces long-term maintenance; DRJTBC's executive director calls it a milestone and the setup a framework for the remaining six facilities.","sledRelevance":"Directly relevant to toll authorities converting to open-road tolling, where AI vehicle classification replaces plaza-based classification. A bi-state U.S. authority with a planned six-facility migration makes this a reference for comparable conversions.","evidence":"Vendor announcement republished by a trade site, with an executive-director quote. Traffic volume and transponder share are given; no classification accuracy, revenue leakage, image-review rates or safety data are reported. Safety and maintenance benefits are vendor claims.","architectureImplications":"Open-road lanes rely on gantry-based transponder reading, image capture and AI classification instead of plaza equipment. Classification accuracy directly affects toll rates by vehicle class, so audit data and manual review paths matter.","governanceImplications":"Toll authorities should require reporting of classification accuracy, disputes and manual review volumes, since misclassification changes customer charges. The migration of six more facilities offers phase gates.","securityPrivacyImplications":"Open-road tolling captures plate images for non-transponder users; retention and access rules are not described.","caveats":"Vendor-authored claims republished on a third-party site; no performance figures for AI classification; no independent evaluation. Original TransCore or DRJTBC releases were not separately inspected.","streamIds":["transportation"],"roles":{"sales":"Interpretation: toll authorities converting to open-road tolling need assurance that AI classification charges customers correctly. A bounded engagement could define classification accuracy, dispute and image-review metrics for acceptance testing across phased facility migrations. This vendor source supports no accuracy, revenue or safety claims. Applicability is strongest for small, high-transponder-share facilities like this one. The six remaining facilities provide natural decision points for any review.","engineering":"Interpretation: AI vehicle classification at open-road gantries replaces axle or loop-based plaza classification, so accuracy must be validated by vehicle class, lane and weather. Engineers should plan acceptance tests against ground-truth samples, monitor classification disputes and confirm integration with the back-office system and E-ZPass interoperability. Proof of value should use the first bridge's data before migrating the other six.","delivery":"Interpretation: delivery is a phased migration across six remaining facilities with an acceptance gate at each. The tolling operations owner should track classification accuracy, manual review rates and customer disputes, and publish results before the next conversion. Acceptance criteria below are proposals, not observed outcomes. Customer service staff should log classification complaints separately so they can be counted. Image retention rules for non-transponder users should be confirmed before the next facility converts."},"roleDetails":{"sales":{"customerProblem":"A toll authority must ensure AI vehicle classification in open-road lanes bills customers correctly as it converts more facilities.","stakeholders":["Toll authority executive director","Tolling operations and revenue staff","Customer service center","Board and bi-state oversight"],"discoveryQuestions":["How is classification accuracy measured at the first converted bridge?","What share of transactions needs manual image review?","How are classification disputes tracked?","What must be proven before the next facility converts?"],"valueHypothesis":"Defined acceptance metrics could reduce revenue and customer-dispute risk across six more conversions; size of benefit unknown.","potentialEngagement":"An acceptance-testing and performance-reporting framework for AI classification across phased toll facility conversions.","unsupportedClaims":["Any classification accuracy level","Safety improvement from open-road tolling at this bridge","Maintenance savings"]},"engineering":{"fit":"Good fit for open-road tolling conversions using AI classification; no performance evidence yet.","architectureIntegration":"Gantry transponder readers, cameras and AI classification feed the tolling back office and E-ZPass interoperability processing.","prerequisites":["Ground-truth classification samples","Back-office integration","E-ZPass interoperability testing"],"constraints":["Bridge geometry and lane counts vary by facility","Weather and lighting affect image classification","Vendor proprietary systems limit independent testing"],"security":"Protect plate images and transaction data with role-based access and defined retention.","proofOfValue":["Classification accuracy by vehicle class against manual samples","Manual review rate per 1,000 transactions","Dispute rate before and after conversion"]},"delivery":{"work":"Run phased conversions with acceptance gates and publish performance results for each facility.","dependencies":["Vendor reporting access","Back-office readiness","Board approval for each phase"],"ownership":"DRJTBC tolling operations, with the vendor operating lane systems under contract.","skillsAdoption":"Staff need skills in image review and classification auditing.","governanceCheckpoints":["Acceptance test before each facility goes live","Board review of first-bridge results","Privacy review of image retention"],"acceptanceCriteria":["Proposed: classification accuracy target met on sampled transactions","Proposed: dispute rate no higher than pre-conversion","Proposed: published results before next conversion"],"risks":["Misclassification charges customers incorrectly","Vendor lock-in across facilities","Data on performance may not be public"]}},"retrievedAt":"2026-11-02T08:34:45-06:00","enrichedAt":"2026-11-02T08:42:10-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Removing toll-plaza infrastructure changes toll collection staffing; the source gives no workforce details.","procurementImplications":"Builds on TransCore's earlier cashless conversion; a single-vendor framework for six more facilities raises lock-in and contract-performance questions.","operatingModelImplications":"Shifts operations from plaza-based collection to back-office transaction processing and image review.","sourceVerification":{"openedUrl":"https://www.fairsonline.org/transcore-launches-open-road-tolling-drjtbcs-new-hope-bridge","referenceExcerpt":"TransCore installed its Infinity Digital Lane System, including VCATS, its AI-powered vehicle classification solution.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}