Sensors provide data that can help municipalities improve things like water quality, traffic flow, air quality, and other quality-of-life issues. Cities are finding that good data supports the effective administration and delivery of services and better living standards—and that the broad application of sensors creates an ample supply of data.
But how much of that data is meaningful?
Clearly, sensors that measure water quality are vital. But placing sensors on every light pole in the name of capturing data is likely to yield diminishing returns. While the cost of data management is decreasing, cities still need to be selective about the use of sensors.
With that in mind, here are six criteria to guide your decision-making when purchasing or placing sensors:
Limit the cost of acquisition as data items multiply.
Set priorities for data sources and types (i.e., health and safety before lifestyle).
Avoid commitments that could escalate costs (i.e., avoid sensors that imply using a mobile ID, or IP address).
Control costs for installation and maintenance (i.e., think twice about installing battery-powered sensors buried into asphalt).
Prepare for ambiguity, as standards are not defined for all sensors (i.e., consider data integration layer to abstract from specialized hardware).
Be judicious with sensitive data. Citizens want to know you are controlling risk and expenses, not them.
In the accompanying graphic, you can identify the “do not trespass point.” Beyond that point, more sensors increase costs while adding little information that’s relevant.
By using these criteria you can analyze vast amounts of relevant data from your internal systems (and a reasonable set of sensors), draw meaningful conclusions, and also report results by data category. The city of Barcelona provides a great example of how this approach was employed.
Getting the most out of your sensor strategy is not about hardware. It’s about leveraging software capable of extracting knowledge out of tons of raw data; software business intelligence, data analytics, digital dashboards, open data, and big data scenarios. These are the fundamentals that should definitely be prioritized over the use of sensors.