With the emergence of IoT and the notion of connected cities, an increasing amount of city services are being connected to the internet. This includes transportation –based assets, such as public bike rental stations. We will outline how to map these stations (PI AF elements), and view any geographical trends which might emerge. Bike share stations are distributed around a city, and provide easy access to a public bike share system. Users can pick up a bike at one station and drop it off at any other station in the network. The bike station status is continuously updated with live metrics such as the number of bikes available at a station (for users to take out bikes) and the number of docks available (for users to return bikes).
This post is based on a summer internship project, undertaken at the OSIsoft Montreal office. For more details, check out "Monitoring Smart City Assets with the PI System"
For more information on how this data was collected and stored in the PI System, please have a look at these blog posts:
- Sending public JSON data to the PI Connector for UFL continuously and silently
- Building a PI AF hierarchy using the PI Connector for UFL
Why use the PI Integrator for ESRI ArcGIS?
The PI Integrator for ESRI ArcGIS allows us to geographically map all of our AF assets. It can then display all the static and dynamic attributes of these PI AF assets in real time. The physical location of our PI AF assets (public bike stations) is incredibly important, as the usage of any bike station will be heavily related to its location. Factors such as proximity to pedestrian malls and office buildings will cause an increase in usage. We could have looked at a station map and tried to manually work out which stations would be favored based on their location, which would have been a very tedious process. A much better way to go about this analysis would be to map our stations, and view our KPIs on a map. This allows us to easily see any geographical trends which might be present.
How does the Integrator for ESRI ArcGIS work?
In a nutshell, the Integrator searches the AF archive for updated attributes, and then can send them to an ESRI online or local portal for mapping purposes. The PI AF database is crawled using the “ESRI Data Relay” which continuously searches for new updates in PI AF attributes. The Integrator setup is hosted on an IIS configuration webpage, where the user can set which PI AF elements and attributes to send through to a map layer. In order to map an element, it must have an attribute denoting its location - this may be an address or a longitude/latitude pair. The map layer, which now contains the location of our PI AF elements along with their attributes, can then be viewed in an ESRI product such as ESRI Online or ESRI operational dashboard.
Analyzing our data with the PI Integrator for ESRI ArcGIS
We started off by mapping our PI AF assets (Bike stations in Montreal) using the ESRI online portal. The examples below show how we can view our elements in a much more intuitive way, and combine secondary data sets such as the location of bike paths, subway stations and bike accidents.
A live, interactive version of the map is available here. Below are some of the individual analyses and displays we can generate:
Mapping Bike stations:
All of our PI AF assets (bike stations) are displayed on an ESRI online map. The PI Integrator for ESRI ArcGIS helps resolve the station's longitude/latitude position to a map layer (LEFT). We can change the marker size to display a KPI of interest, such as the total in/out traffic events in a four hour period (RIGHT).
Zooming in on an area of high use, we can take a look at station closer to Montreal's OSIsoft office. We can see all of our PI AF attributes displayed in real time, updated at an interval which we've specified (5 minutes).
We can further enrich this data by adding other map layers. For example, we see that bike stations with the most usage coincide with the location of bicycle paths (LEFT, in purple) and the location of subway stations (RIGHT)
System administrators might want to compare this year's live data with last year's historical data to see which stations are under-performing and which are exceeding last year's usage. In this case, we can plot last year's trips in red (LEFT) , and then overlay this year's live data in orange (RIGHT). We see that for the most part, stations this year were more popular than those last year.
Another dataset which city planners may be interested in is the location of bike accidents in Montreal. This would help point out any particularly dangerous intersections, and lead officials to investigate whether a bike dock should be placed in close proximity to them. Worryingly, there is much overlap between high use bike stations and intersections with a concentration of bike accidents.
Given that ESRI is a worldwide mapping application, we can easily move to a different city and view a bike station layer which was generated. For example, if we wanted to have a look at bike stations outside of OSIsoft’s Philadelphia office (1700 Market St.)
In addition to Philadelphia, we are gathering data from other bike sharing systems such as New York City’s CITIBike, San Francisco’s Bay Area Bike Share, Boston’s Hubway, and Toronto’s City Bike Share. If you were to find yourself in Times Square, you could consult an ESRI map to see live PI AF attributes from New York’s public bike sharing system.
For those who are reading this in the Bay Area, we can also have a look at San Francisco’s Bay Area Bike Share.