FourKites shares insights on best transport management practices

Air, sea, and surface transportation/logistics modes will be examined closely by panel of industry experts when the 2019 USC Marshal Global Supply Chain Excellence Summit convenes in August. Among the speakers will be Mathew Joseph Elenjickal , Founder and CEO, FourKites. In this exclusive interview, he shares a few insights on the current state on best management practices. 

Supply Chain Management Review: How does predictive analytics mitigate risk for supply chain managers?

Mathew Joseph Elenjickal: Transportation management used to be a very static and reactive process - managers would be stuck trying to make amends for a late load after it’s already missed the appointment and disrupted processes downstream. But when AI and machine-learning methods are applied to a vast pool or transportation data, like the one FourKites collects, managers can start to be more proactive and even prescriptive in their planning. They can make better decisions based on trends in the data analytics, and steer clear of future bottlenecks that they receive alerts on - even before a load has left the facility.

SCMR: Any examples? 

Elenjickal: Yes, for instance, you can build your transportation plan based on billions of historical data points, and determine the best ways to optimize your end-to-end supply chain. Machine learning can then tell you which lanes, schedules and facilities will help you get your goods in on time and in full.

SCMR: And if something does go wrong?

Elenjickal: Predictive analytics can help manage exceptions or minimize the impact to your operations. If a driver is delayed at the warehouse or distribution center, you’ll be able to see an updated predictive ETA for that shipment and take corrective action before your customer is impacted by the delay.

SCMR: Where is this ongoing evolution headed?

Elenjickal: The platforms with the densest data networks will be the most impactful when it comes to revolutionizing supply chains as we know them. In the supply chain visibility space, for instance, a robust data network allows for a continuum from reactive (“Where’s my truck?”) to proactive (“When will my trucks be there?”) to predictive (“What are the risks to my loads, and what can I do to avoid them?”) to prescriptive (“These are the risks to my loads, and they will be fixed automatically”). This functional evolution is only possible for those platforms that are amassing huge quantities of data each and every day.