National Rail Enquiries, a company that provides ticketing and travel information to customers of train-operating companies, has gone out to tender for a new interactive voice response service.
The voice service, dubbed TrainTracker, allows customers to request and receive real-time train-running information, including full journey planning up to three months in advance and information about the cheapest fares available.
The technology is currently provided by BT, which will be invited to tender – along with other providers – before the current contract expires in September 2012. To ensure a replacement technology is implemented by this point, National Rail Enquiries intends to select the winning vendor by the end of March 2012.
The tender notice reads: "The updated TrainTracker service is expected to use industry leading voice recognition technology capable of accurately and consistently recognising natural language and multiple item requests in a single utterance.
"The service will interact with [National Rail Enquiry] APIs for the provision of train running information and is expected to include self service elements to enable control and manipulation of reference data and call flow."
National Rail Enquiries has also indicated that it will use service-level agreement metrics, including customer satisfaction, which will be determined by automated surveys, and an overall service success rate.
Andrew Tolley, real time systems manager at National Rail Enquiries, told Computing that the new technology would be suited to a cloud-based product due to the fluctuating nature of the call patterns.
"Our volumes vary between well-known and controlled, and volumes that increase with little or no warning," said Tolley.
"We would absolutely consider cloud services, particularly as we have very short-term, unpredictable peaks in usage due to things like bad weather and train strikes."
On average, National Rail Enquiries receives 6,000 calls a day, which equates to an estimate of 15,000 call minutes.
However, this can increase by as much as fourteen times to 84,000 calls a day during a period of disruption.
"Cloud would give us the ability to go in and deal with a sudden and unexpected volume of calls, which would suit our services very well," said Tolley.
"It is certainly something that we are interested in exploring further."
By eliminating high entry costs for big data analysis, you can convert more raw data into valuable business insight.
A discussion of the "risk perception gap", its implications and how it can be closed