Better mobility options: Bike sharing with big data

IN General — 18 June, 2014

Particularly in large urban areas, the automobile is no longer the only real mobility option. But whether car or bike sharing, the same applies: for the sharing providers, the partial models on offer are problematic in terms of logistics and service. In a model analysis, Blue Yonder has demonstrated that using predictive analytics bike fleets can be accurately managed and customer satisfaction can be increased.



The lower the bar, the more accurate the prediction. Blue Yonder technology clearly reduces the inaccuracies of forecasts. The lower the bar, the more accurate the prediction. Blue Yonder technology clearly reduces the inaccuracies of forecasts.


In Berlin, a city that loves bikes, a local newspaper, "Tagesspiegel" discovered a new development that is a hot topic online: the digital bike lock, which can be operated simply, by smartphone. The digital bike lock opens new horizons for bike sharing, which is considered the little brother of car sharing. The following is true of both models: The more flexibly customers can get on a bike or behind the wheel of a car, the larger will be the market for new mobility concepts.

The German Bundesverband Carsharing (The German Federal Carsharing Association) finds that station-independent providers will be able to double their user figures within a year. The German newspaper "Handelsblatt"  has the figures: There are now 437,000 people in cars in Germany that can be left almost anywhere. In total, the German car sharing market has 757,000 people behind the wheel. With near 14,000 vehicles, this means that there are 50 users for each vehicle. How can vehicle sharing concepts be realized in such a way that the customer is continually and everywhere mobile, but at the same time, the provider can avoid over-capacities?

Weather? Day? Time? Not all factors effect demand

Blue Yonder addressed this issue in an exemplary way for bike sharing. With a demand forecast based on predictive analytics, providers can align their logistics and

services much more closely to what the users want than was possible with traditional prediction methods. We analyzed data from 2011 and 2012, for a total of 17,000 hours. Data on the weather, days of the week, and seasonal and daily data was included. Using data analysis based on our algorithm, we were able to quickly find out how the various factors effect bike demand, and which factors don't play any role at all. In this way, we forecasted the number of rented bikes on an hourly basis.

What advantages does big data analytics provide to the market? Bike share providers can position their bikes in those places where the bikes are needed at a specific time. Bike maintenance can be planned in such a way that there are no bottlenecks. Customer satisfaction increases with growing availability. And the more people who peddle, the more complex bike sharing will become as a model for the future. For that reason, it is a good idea for providers to deal with the subject of automated demand forecasts early on. The "pioneers" —the auto lessors and manufacturers —are a classic example of automated mass decision-making, with their car sharing offers.

Blue Yonder Blue Yonder

We enable retailers, consumer products and other companies to take a transformative approach to their core processes, automating complex decisions that deliver higher profits and customer value using artificial intelligence (AI).