London courier firm eCourier has gone live with an intelligent software-based service that can automatically allocate couriers to current jobs.
The company has also rolled out 3D mapping software that allows customers to track their deliveries in real time via the web.
eCourier manages 80 courier vehicles, including motorbikes, bicycles and vans, each of which is equipped with an O2 XDA II handheld device running Windows Mobile 2003. They also have GPS modules that communicate each unit's location and availability back to a main server every 10 seconds.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
The IT system has helped eCourier win business from companies including Harrods.
"The new auto allocation component provides the customer with estimated delivery times and allows us to run an end-to-end automated business via the website," said Jay Bregman, eCourier co-founder and chief technical officer.
He added that customers could add deadlines to their deliveries via the website, and this becomes part of the auto-allocation calculations.
The main difficulty with the auto-allocation system was tuning it, said Bregman.
"We put in all the rules, but mimicking the actions of a human controller is such a complex process; the way the rules interact do not always come out the way you want it," he said.
The system is based on a SQL Server 2000 database holding customer data, two dual-processor auto-allocation servers running Red Hat Linux and a MySQL 5 database.
This acts as a neural network, forecasting things such as travel times, weather and traffic. One of the Linux machines is a high-performance Intel server with 64-bit extensions running Red Hat Enterprise 4.
"We chose Linux because of its reliability - the auto-allocation is central to our operations. Red Hat Linux with MySQL is fast and cost effective," said Bregman.
eCourier also uses a heavily customised version of Salesforce Enterprise Edition to capture customer data and give it a unified view of its customers.