How Data-Driven Decision Making Changes The Way We Operate
Blog

How Data-Driven Decision Making Changes The Way We Operate

By Jasper Neef on

For decades, in the geospatial industry, we have relied on systems that present mostly historical data -whether it is data from last week’s inspection or base data collected years ago. On many occasions, these various data sources are used in unison. As a result, decisions based on these systems rely mostly on mathematical models, attempting to distil a reliable view of reality.

With the rise of concepts such as Internet of Things, access to real-time data is growing by the day. On a professional level, sensors measuring everything from water quality, to waste bins’ fill-levels, and noise and air pollution, allows our customers to be on top of everyday operations like never before.

On the flip side of this, a new problem arises. With every new sensor added, more data is being collected that needs analysis. Data, in itself doesn’t necessarily hold value, rather what it represents ultimately counts. Every human being can only process so much data in a meaningful way. And with terabytes of new data coming in day by day, we are pushing our limits. Especially at the managerial level, decisions have to be made rather quickly, and without the luxury of deep diving into the raw data. What starts out as a blessing can quickly turn into a hurdle for decision making, that is, if we are unable to draw the essence from the data deluge.

The trend of using sensors to capture real-time data is irreversible, and rightfully so. The key to using this data effectively is to sift through all measurements and focus on trends and irregularities. This requires new evaluation methods and ways of communication. An effective example of the former is the upcoming use of Machine Learning for assessing situational information. By comparing a given situation to hundreds or even thousands of similar and known situations, the assessment of what’s going on can be made much more effectively than by testing it against a given set of rules.

To put this into a practical context, we can use the example of detecting unwanted obstructions in waterways. IMAGEM has developed a machine learning model that analyses long stretches of waterways and compares possible obstructed locations against a set of known obstructions. This has vastly improved the detection results and in turn, allows the waterboard to much more efficiently deploy their surveyors, saving a lot of money.

Another example of using automated assessment of raw data is the detection of water permeability based on infrared aerial imagery. This analysis provides a good indication to the public works department if problems are likely to occur in certain neighbourhoods when heavy rainfall is expected. By subsidising the removal of pavement in gardens, that problem may be reduced.

Where it comes to communication, traditional GIS portals are giving way to location-intelligent dashboards, that not only visualise where a certain phenomenon occurs, but also put it in context by assessing its impact on policy goals. This is the information policy makers and managers need to make informed decisions. For example, by analysing air quality levels over a period of time, and confront these with quality goals, a decision maker knows when and where to take measures. When combined with real-time traffic information and socio-economic parameters per neighbourhood, the complexity of the problem is explained and options for improvement start to arise.

Smart M.Apps do just that. By processing new data automatically and aggregating it in concise dashboards, they put the tools in the hands of policy makers to quickly react and improve the liveability of their town or city, without them having to do the fundamental analysis. The complex interpretation and processing is done in the background, and only the very essence of the problem is presented or visualised. The combination of new complex data interpretation through Machine Learning and intelligent dashboarding allows users of Smart M.Apps to focus on problems, rather than data.

Author

Patrick de Groot, Business Development & Sales Operations Manager at IMAGEM

Patrick de Groot is the Business Development & Sales Operations Manager at IMAGEM. He has over 18 years of experience in business development in the geo-ICT industry. At IMAGEM he has held multiple roles to help the company strategically grow their business and revenue. Currently, he leads a team of sales and account managers and business consultants focusing on increasing market footprint in multiple industries. His background in geospatial technology gives him an edge and enable him to help resolve problems faced by customers. Patrick has a bachelor’s degree in Town & Country Planning and a degree in Business Administration. Outside of work, Patrick is an avid bicyclist and enjoys camping with his family.

Connect with Patrick on LinkedIn and Twitter.