- The Problem
Driving apps and maps have revolutionized the way we get around our cities. The widespread adoption of Waze, Google Maps and other routing apps demonstrates that they meet a strong need in the driving community. These routing apps come with consequences though. They have routed large volumes of various types vehicles - from cars to freight - along residential streets and work & school zones, creating safety concerns and eroding the quality of life on those streets for which they were never designed for.
While their intent is to reduce single occupant vehicle travel times, driver apps can create the congestion that they are trying to avoid, they route too many vehicles along a single route when there are other equally favored routes.
These problems all stem from the fact that routing apps only optimize for individual trips and they do not know about city priorities and restrictions. They do not consider the concerns of those living on these residential streets that make up the routes, and they do not consider the effects of these trips on the traffic network as a whole.
- The Solution
Other than adding more lanes or roads, we propose making better use of the roads you have. For the City of Pittsburgh we propose a holistic approach to routing equalizing and traffic and which gives city and its citizens a tool for influencing the routes generated by routing apps in and around the city. The routing engine generates a weighting factor for each road segment in the city based on the traffic conditions, type of road and neighborhood or other city priorities and restrictions. The higher a segment’s factor, the more routes are discouraged along that segment. These weighting factors are updated once a minute and pushed out to all integrated routing apps.
We call this a Collaborative Routing Engine (CRE), a fully automated solution that does not require any operator intervention during normal operations. Traffic managers configure the rules which drive the scenarios, and the engine takes it from there. The engine may be integrated directly with existing operational and data systems, if available, and use those data to drive the process. The engine is a general-purpose tool which will solve daily commute issues such as:
- Citizen Concerns: Routing apps have created problems by sending commuters down residential streets in large numbers. The city can use the CRE to dynamically assign priorities to road segments so that routing apps avoid sending vehicles through residential areas, school zones or near hospitals. This approach allows some through traffic on residential streets, but limits it. In other cases, such as school zones, the CRE can reduce essentially all commuter traffic during school arrival and dismissal hours. The CRE gives the city a powerful tool for affecting the way its streets are used.
- Better use alternative routes to avoid congestion: Where there are multiple routes between two destinations, the CRE can distribute the traffic across those routes by varying the weighting factors over time. At each successive time interval, a different alternative will be favored, and the traffic will be spread out. This is fundamentally different from what routing apps do not in that routing apps send everyone along the same route until that route gets clogged and slows down. Only then do routing apps move travelers to other routes. CRE can distribute traffic before the slowdowns begin and prevent jams from happening.
- Incident Response: It can be integrated with an incident management system, then it can update weighting factors and begin routing traffic around an incident before traffic slows down. Currently routing apps only route drivers around crashes once traffic stops. The CRE can redirect traffic immediately so almost nobody gets stuck.
- Traffic Predictions: Kapsch has a predictive traffic system which could be integrated into the CRE in the future. The predictive engine uses advanced artificial intelligence to estimate traffic volumes up to 45 minutes in the future. Using this data, the CRE can route vehicles around congestion before it happens.
The CRE is a win-win proposition. By balancing traffic across the whole network, drivers get overall shorter commutes. Residential citizens have a way to limit through traffic in their neighborhood, and the routing applications get a solution which addresses some strategic shortcomings in their products. The City Traffic Manager has a way to manage demand through load balancing and a way to encourage adherence to city priorities and restrictions.
Integration with existing routing applications is the most powerful approach but may not be feasible in a pilot project. The engine will be delivered with an API which can connect to 3rd party navigation apps. To initially test this out, Kapsch will also deploy a routing app for the initial pilot. This app can be made public and promoted as needed for the purposes of the pilot. If Pittsburgh has variable traffic signs they can show alternate routes and travel times via available signs and the apps.