Advancements in mobile networks and near-ubiquitous smartphone use have propelled location-based services (LBS) into prominence. The geopositioning ability of smartphones, combined with widespread mobile broadband, enables firms to collect geolocation data in near real-time and subsequently use this data to inform a range of services.

Take, for example, Uber – the ridesharing firm continually improves its products with data-driven routing and pricing optimizations, which are prime examples of how geolocation data can be leveraged over mobile networks for service improvement. In December 2016, Uber announced Automatic Trip Upgrades and Predictive Matching for its ridesharing service uberPOOL – the former feature uses real-time data to ensure that ridesharing matches preserve route optimality, while the latter uses historical data to make the most optimal ridesharing match when traveling along a route[i]. In late 2016, Uber also expanded its app’s data collection capabilities, allowing it to access a rider’s location data from the moment a ride is requested to five minutes after drop-off. Uber aims to use this additional data to address imperfections in the pick-up and drop-off processes[ii], such as identifying potential safety hazards by detecting when riders cross a street after drop-off. Geolocation data is even transforming the wider transportation industry – a recent MIT study found that by introducing ridesharing and directing drivers to high-demand areas, just over 20% of the current taxi fleet could meet 98% of demand in NYC[iii].

> Click to interact on your mobile device: NYC Uber Pick-Ups: Trends During the Workweek

We analyzed some of Uber’s publicly available geolocation data[iv] and visualized pick-up locations in NYC across a typical workweek, revealing time-based usage trends and geographic hotspots.

Communications service providers (CSPs) enable these processes through the connectivity services that they offer. For instance, to implement Automatic Trip Upgrades, Uber must receive data from the mobile devices of all active riders and drivers in a given area, run an algorithm to find optimal ridesharing matches, and finally transmit notice of service changes to affected users – all in near real-time. Furthermore, the ability of LBS to proliferate and scale is similarly dependent on CSPs – specifically, on the level of coverage and quality of service provided by the mobile networks that they operate. Uber now handles over 215,000 rides per day in NYC[v], with demand fluctuating based on time of day and geographic location as seen above. To accommodate the variable demand generated from Uber and other LBS, CSPs may inevitably need to invest in small cell deployments or other network redundancy infrastructure to boost coverage, especially in high-usage areas. Additionally, quality of service enhancements can help LBS operators achieve incremental service improvements. Reduced latency and faster data transfer rates can boost the efficacy of the Assisted GPS (A-GPS) often used by smartphones by further reducing time-to-first-fix (i.e. initial GPS signal lock). Dense small cell or carrier Wi-Fi deployments can increase geopositioning accuracy, as the presence of additional cell sites and Wi-Fi access points allows A-GPS to augment GPS location with cell site multilateration and/or Wi-Fi positioning.

The advent of 5G mobile networks, which promise greater coverage, higher speeds, and lower latency, will further enable the proliferation of LBS and empower firms to collect more granular geolocation data. Whether to better target advertising campaigns, offer personalized promotions, or enhance a virtual reality experience, more and more companies will leverage geolocation data for service improvement. Ultimately, the level of insight and value that can be derived from location-based services is tied to the rate at which communications service providers continue to grow and innovate their mobile network capabilities.


Sources:

[i] https://newsroom.uber.com/effortlesspool/

[ii] https://techcrunch.com/2016/11/28/uber-background-location-data-collection/

[iii] https://www.csail.mit.edu/ridesharing_reduces_traffic_300_percent

[iv] https://github.com/fivethirtyeight/uber-tlc-foil-response

[v] http://toddwschneider.com/posts/taxi-uber-lyft-usage-new-york-city/