Better forecasting is here now with a hybrid cloud approach

By | 2 minute read | December 15, 2020

bike stand

This article was originally posted as Better forecasting is here: Get to know IBM’s Hybrid Cloud strategy by Filippo Focacci featured on the IBM Journey to AI blog on October 23, 2020.

Whether inclement weather or a wildly fluctuating economy, planning for the future is heavily dependent upon understanding a complex interplay of factors. Every business craves a crystal ball, but to predict and shape future outcomes, enterprises need to successfully collect, organize and analyze their data—a challenge most companies struggle to do or do well.

Modernize your data: Clearing hurdles to AI

DecisionBrain has long had a unique relationship with IBM. As a former IBMer, I worked in the ILOG for 15 years before leaving to start my own venture and DecisionBrain’s employees boast global expertise in IBM Data and AI solutions, with some having also worked for Big Blue.

Our partnership with IBM has offered us a clear path to success with its hybrid cloud strategy. At its core is IBM Cloud Pak for Data, an information architecture for AI that automates the often complex and messy steps needed to get to AI successfully.

Keeping London’s Cycle Hire scheme humming with Cloud Pak for Data

Take our work with Serco, a company that operates a bike-sharing service throughout London. The organization wanted the most efficient and cost-effective way to manage and maintain 12,000 shared bicycles across 800 stations throughout London.

For this project, we needed to analyze the bicycles usage patterns, so we deployed a predictive and prescriptive solution to calculate the optimal number of bikes needed at each station at any given time. We also needed to plan the most efficient routes for maintenance teams to repair and redistribute bikes accordingly.

After training the solution on two years of historical data, the tool was ready to provide real-time predictions for where bikes will be needed, and to provide real-time recommendations on how to get them there, every day. The solution runs online and is available 24/7. The results were seen in a decrease in company costs and a more efficient bike sharing service overall.

Today’s predictive models forecast the number of bike pick-ups and drop-offs of each station throughout the day. The model considers circumstantial data, such as historical bike demand, historical weather data and current forecasts, to make a unique prediction for every day.

Watch this video for a closer look at DecisionBrain’s impact on Serco’s bike sharing program in London:

As a business partner, one of the biggest advantages of Cloud Pak for Data is that it brings everything we need into one single place, seamlessly collecting, organizing and analyzing data regardless of its type or where it lives. This makes our operations much easier and reduces the cost of building our projects. We have predictive capabilities, decision optimization and machine learning technology all in one centralized location, from which we can easily deploy to any cloud.

IBM Cloud Pak for Data can help you scale the value of your clients’ data to help them become a more predictive enterprise.