Using Google Maps Tech For Liability Management in Public Works Departments

SpringML was recently awarded the status of being the Top Tier Partner for solutions requiring Google Maps Expertise. We were glad to leverage our knowledge and understanding in the domain to solve customer concerns around liability management, specifically for Public Works departments.

For example, SpringML used machine learning models and Google Maps to help the city of Costa Mesa by saving $40,000 each year and freeing up money and labor for other critical projects.

In this Blog we are going to understand how we used a universally accessible tool, like Google Maps, to solve concerns around infrastructure management and liability reduction for various customers and how it can prove effective to solve similar concerns in the future.

With Google Maps Platform’s latest Maps, Routes, and Places features, one can create real-world, real-time experiences. Many services are available on the Google Maps Platform, and the following are employed in our current use cases. These APIs can be used by just enabling them in the GCP console.

Using the Google Maps Javascript API, one can embed maps in our web applications. Google Maps also has a Geocoding API that can be used to acquire the coordinates (latitude and longitude) for a given address and the other way around.

City of Memphis:

We leveraged Google Maps to visualize the locations within the City of Memphis where potholes are detected. The geo-coordinates are stored for each pothole detected. To detect the geo-coordinates in the frame, Google Vision API was used.

The locations are saved in Google Cloud’s data warehouse BigQuery in the form of latitude and longitude. Using Google Maps API, the latitude and longitude are converted into an address using reverse geocoding.

Upon validation of a pothole by a user through the web app, a request is sent to a series of web services operated by the city as well as the Google Maps API. These will convert the detection metadata into a consumable format for 311 and will then submit a ticket to the system. From there a maintenance crew can plan and respond to issues with more information than they had before.

Costa Mesa Sanitary District:

For Costa Mesa Sanitary District in California, we inventoried manhole covers around the city and detected damages around the apron. The city wanted to track which manhole covers around the City have more damages around them so they can be tracked and fixed as a part of proactive maintenance. We were able to detect the manhole covers by driving around the city using a smart camera which captured the video footage as well as location details. Upon detection, the geo-coordinates of the manhole cover were extracted using metadata and pinned to the embedded google maps.

Each manhole cover was represented on Google Maps using a color based bin. Each rating was represented with a different color pin. Using Geo-coordinates., Google Maps API was also used to reverse geo-encoding the coordinates to the addresses so that it becomes easier for the county to track them.

Map View:- Each pin represents a specific rating of the manhole cover.

List View: Using embedded maps to trace the location of each manhole cover. Using the google street view embedded in the map, we were able to cross verify the location of the manhole cover.

The artificial intelligence augmented with the Vision AI Platform and Google Maps Expertise can help organisations to better understand, plan and respond to various infrastructure and transit-related use cases. The end-to-end platform, mentioned above for the customer use cases could be repurposed for various other liability management, research and cost management projects. This same technology could help in areas like transportation/traffic, sewage, land, waterworks etc. to identify plausible solutions to different infrastructure management concerns.

This technology and the robust analytics system provided alongside, could also be used for identifying liability/safety concerns in transit spaces like Bus Stops, Subway, Railway Stations, Airports etc. and also identify footfall in these areas to understand how these places could be used for retail and advertising purposes.