How has Predictive & Perceptive Analytics Transformed Location Intelligence?

 

– Bishwajeet Kumar, Tanaya Jawale, and Vishwanath Rajput

 

In Christopher Nolan’s blockbuster thriller “The Dark Knight”, there’s a scene where the Joker plays various sadistic games with the citizens of Gotham. Batman and his colleagues track him by using an unexplained technology to convert cell phones into a city-wide tracking surveillance system which took advantage of the phone’s GPS locators. Each movement could be tracked and was acted upon.

 

Today, that vision of the future is coming closer to reality. Businesses are capturing vast quantities of data with the end goal of giving you more of what you want, when and where you need it (Of course not by Batman himself). It’s the ultimate consumer experience. The simplicity and ease of use GPS-enabled cell phones and mobile devices help you find the closest place of interest. Maps let you visualize where you’re going and where you want to be. All those cool gadgets are no longer exclusive to Star Wars.

 

Location Intelligence & Analytics has now become a core data source that helps in contextualizing of location-centric data, so that meaningful insights could be derived from the same and strategic business decisions could be made. It has emerged as one of the major segment in the field of market research.

 

Having stated the obvious necessities for tracking location, lets’ focus on how it is being currently done and “Are we doing it efficiently?”

 

As of now, GPS-enabled cell-phones and mobile devices form the major source of raw location data. Given these individual real-time sources, and possibly additional information available at a location via static sensors, it is possible to monitor and track what is happening, what has happened and possibly predict what may happen at every location. There are predictive algorithms which can even track the leaning angle of an object (required for marine buoys), altitude, etc. Google’s Places and Locations API can provide Location Category, Price-Level Category, Geo-fences, etc. Using these predictive & perceptive algorithms the category and amount of information that could be retrieve is tremendous and can provide critical insights for any business model.

 

Download: Whitepaper on Industry Trends in Field Service Optimization

 

The major challenges faced while implementing efficient location tracking ranges from battery drain to network signal reception to optimizing time and distance based location updates. One must dig-deeper and optimize their location tracking implementation depending upon their model and use-case. Models involving near real-time location updates, like Uber, would have to maintain a balance between battery drain and location update calls. One major drawback with location based apps is that even the location services are optimized, the product model requires that data to be processed and meaningful information be extracted for them to function. Geocoding and Reverse-Geocoding libraries are yet to be perfected. Open-Source community has come up with various alternatives for Google’s Geo-Coding Library but much work is still to be done.

 

There are various questions a brand or business entity needs to ask their tech partners regarding their location intelligence & analytics before acting upon it:

 

  • How is the location data sourced?
  • What data is unique for their business model?
  • What 3rd Party dependencies are involved?
  • Is the end result data is inferred or extracted?
  • How accurate is the end result?

 

So, what’s next?

 

According to Gartner Reports, Location Intelligence & Analytics is still an evolving technology which with the advent of AI and Machine Learning will evolve drastically. Recently, Google released Tenser Flow, an Open-Source, Cloud based software library for machine learning. Tenser Flow along with Neural Networks will provide ample boost to this technology for business implementations, various IoT devices like Motorola Connect will take it to households.

 

Each of these technological needs is pushing the technology envelope in different areas:

 

  • Spatio-temporal modeling, databases and warehouses.
  • Handling raw data and processing these event streams at scale.
  • Monitoring and detecting trends – pushing the frontiers of statistical modeling, artificial intelligence and machine learning.
  • Algorithms to process streaming data to characterize the underlying phenomena.
  • Mobility modeling of entities and other locational phenomena.

 

Utility for Location Intelligence & Analytics ranges from as primitive as Ant tracking algorithm for surface discontinuity extraction-faults detection to huge marine vessels’ Global Ship Tracking System.

 

Field Service Management Future Trends

♥201 135 Subscribe