Case Study: Directed Technologies rethinks scale and data

Automotive, transport and logistics solutions company, Directed Technologies transformed its monolithic database into a microservices architecture following a reshaping of its approach around data and scale.
 
Brent Stafford, executive director at Directed Technologies told Digital Nation the company saw a “massive increase in scale”, beyond standard road deliveries but also a “rapid rise in the smaller vehicles, micro delivery vehicles.”
 
From here, Stafford said this led the team to rethink its view on data and scale as the number of assets and data types to analyse increased. 
 
“We had this monolithic database design that needed to be transformed into a microservices architecture,” Stafford said. 
 
“We needed to migrate more to a data lake approach. To do that, we did an extensive scour of the world's IoT platforms and cloud services and database platforms. 
 
“After an extensive review process, we landed on AWS and Single Store, allowing us to transform the way we think about data and the way we can ingest and process and transform and analyse that data,” he said.
 
Today, Stafford said with the application of AI, machine learning and neural networks, it's critical that there's access to a high-velocity pipeline of data.
 
“In our world, that might mean that there's a vehicle rollover or a driver duress button has been pushed and we have to immediately notify emergency services and fleet managers. We might need to detect things at the edge. 
 
“For example, the truck has seen a stop sign and we want to detect that at the edge quite rapidly. 
 
He explained processes such as neural networks and the high velocity of processing had to be conducted both within the vehicle and also in the cloud for subsequent data retrieval and processing. 
 
“That was vital to provide the analytical insights that the manufacturer, the truck dealer, the fleet manager and even the driver needs to ensure they've got a safe and efficient operation of their fleet or their daily drive,” Stafford said. 
 
The overall implementation was a five-year project according to Stafford, with the build taking about three years.
 
“Today we're applying quite sophisticated algorithms and learning models onto the data,” Stafford said.
 
He added, “It was quite important that the organisation could get behind one platform instead of having to learn multiple platforms.” 
Stafford said customers' reactions to the changes have been positive, as they too can now gain insights into different parts of the organisation that used to be siloed across various managerial positions. 
 
“To cut through and not provide data overload but overlaying those analytical insights allows the fleet manager to get and act upon the information that they need at the time they need to act,” Stafford said.  
 
“Other information that may not be in their direct wheelhouse can go to HR or finance or admin and ease their workload so they can have a more efficient fleet operation,” he said. 
 
The high-velocity data processing pipeline also brings the company new opportunities. 
 
“We look forward to some really exciting areas that this high-velocity data processing pipeline, with things like neural networks at the edge … right through to the cloud services, where we can look at those long-term series analyses to look at optimising fleets or optimising the manufacturer of the vehicle journey as they build,” Stafford said. 
 
Directed Technologies is focussing on local implementation as the business “can’t always compete at the global scale.”
 
He said the role the company best serves is the digital transformations of regional businesses plus also helping organisations embrace a more holistic single platform view of data.