• Many municipal public works agencies know the benefits of having an automated vehicle location (AVL) system to monitor employee performance, improve operational efficiency, and increase public communication efforts. But another little-known benefit is being able to map the discrete locations of potholes or other “points of interest.” Once such points are identified, a maintenance plan can be put in place.

    Any AVL system includes:

    A geographic positioning system (GPS) device

  • A wireless network
  • Back-end system software including a database and geographic information system (GIS) maps.

Most commercially available GPS devices have discrete input lines to monitor the state of the lights, sirens, plow position, or any other on/off system in the vehicle. Two to eight input lines detect voltage changes. For instance, when the emergency lights are on, a specific line senses voltage. When the lights are off, no voltage exists. This simple logic allows an AVL system to monitor the position and state of various devices.

The back-end system software stores and displays the AVL information. Dispatchers or supervisors monitor the fleet with mapping software, the most visible information. However, all information collected by an AVL system is stored in a relational database. This data can be used to replay events, generate fleet analysis reports, and archive historical AVL trends. A common database schema would look like a spreadsheet such as this.

In the previous example, the last four columns show zeros or ones to indicate the states of the inputs. In1 may signal ignition on/off, In2 may signal plow up/down, In3 may signal spreader on/off, and In4 is unused. This unused input line may be used to map potholes.

Pothole Button

A simple momentary switch (a push button) can be connected to an unused input line and mounted near the dashboard. When the driver passes over a pothole he or she simply presses the button to generate a standard AVL message containing a one (1). Using the previous example, the database then would look like this.

This record also includes the latitude and longitude of an event that corresponds to In4 being activated (a button push) with accuracy better than 30 feet most of the time. Presuming the operator presses the button only when driving over potholes, one can mine the database for records where In4 = 1 and map the corresponding latitude and longitude in the GIS. The result is a map of the potholes. Also, because the precise time of the button press is stored, it is relatively easy to classify potholes based on the time they are discovered. Potholes older than two weeks should be repaired right away, whereas potholes discovered just yesterday can be scheduled later, based on a maintenance plan.

Mapping Requirements

While the logic behind pothole mapping using AVL is quite simple, some requirements must be met. For one, the system must support discrete input monitoring (it senses the state of a system by the absence or presence of voltage) for the pothole button to be connected. Fortunately, this is common. In addition, for data mining, an agency's AVL database must be accessible. AVL systems hosted offsite may not allow their customers direct access to the tables containing the actual data. There is something to be said for keeping the AVL system in your closet. An agency will also need knowledgeable GIS personnel to facilitate the data mining and GIS plotting. Finally, the agency must have a maintenance plan. Mapping potholes but not fixing them will provide few benefits to the agency or its customers.

Using AVL to map pothole locations is a great way to extend the benefits of a system. It is an inexpensive way to collect data that would otherwise be labor-intensive. Municipal vehicles drive up and down the roads every day, so adding a pothole button and training drivers to use it is simple. Combining this with a maintenance plan can help to ensure road safety and will certainly improve customer sentiment.

— J.D. Main is engineering director of CompassCom Software Corp., Centennial, Colo.