A Fully Digital Service
Background
Voice Resource Management (VRM) was a manually performed process, from order to allocation and further provision. Circa 20 large scale, complex orders were made every month, each one taking around 12 days to complete.
To effect most orders, at least 15 systems were involved in the provisioning process, many of which were legacy systems. They were hard work to use, inconsistent in performance and were somewhat fragile given their age. They were not designed for the demands of a modern CSP. Further issues existed due to inventory system which was not fully digitised for the provisioned services.
The architecture of the VRM was very manual in nature. The process was inconsistent and this led to lots of mistakes, itself leading to time wasted, lots of review and stifled quality assurance. The legacy systems within the process were already out of support posing a risk to the company and the overall business. Additionally, the inventory system was not easily accessible by the rest of the business causing operational issues.
Objectives
Deciding to automate the process, the CSP faced obvious challenges so a legacy friendly automation capability was necessary. CORTEX is able to operate with legacy systems, and appears to offer more compatibility that other tools available today.
To test the CORTEX capability, the project was then broken into phases. The initial phase was to test by simply starting the allocation and provisioning of a new service B2B service, within the existing infrastructure. As confidence built, this was extended to include B2B on new infrastructure so the customer had a consistent set of tools and processes enterprise wide.
Having recognised the CORTEX capability for legacy technologies, the customer continued its iterative investment then choosing to replace the legacy BSS stack, helping existing business methods integrate into the new process and new tools.
Finally, the customer’s Product team wanted to deliver a fully automated solution to enable an end-to-end digital experience for their B2C/end users which posed a volumetric challenge for the operator. Unlike the high complexity, low volume automations already tried and tested, this challenge was one of scale; high volume, but relatively simple automation. This was contrasted against modern RPA capability from another vendor to CORTEX.
The end vision for the business was to provide a self-service product for their customers be they B2B or B2C, from end user online purchase to delivery within minutes.
Solution
CORTEX gathered order information from the Technical Order Management (TOM) system for the provide, amend, and cease functions. The orders were decomposed, and each part of the service was allocated in the relevant network system.
The allocation was captured in the new inventory and then provisioned through the reuse of existing low level multi-vendor task automation into the network devices.
During the initial delivery, the business requested several manual validation stages whilst they adapted to the automation. This was delivered very quickly, and early value built confidence. As confidence grew in the platform these checks were gradually removed, increasing the value and speed of execution.
As part of the delivery, the legacy BSS stack was replaced with a strategic one, that was automation ready. The encapsulation capability within CORTEX, alongside extensive integration features, allowed a straightforward parallel execution between the legacy and modern systems, ultimately removing the need to interact with the legacy systems entirely, and some systems were then able to be removed.
The CORTEX automation solution delivered a user-facing UI, as requested, to visually track the automated steps, and so in the earlier stages, the customer could manually intervene to resolve exceptions and audit the automation. These were requested all across the business. Human-in-the-loop capabilities are a normal request when automation is wide reaching and a significant change for an operation. The advantage of CORTEX meant the diverse range of requested UIs could be deployed and controlled centrally, being retired as and when appropriate so alternative dependencies didn't build.
Outcomes
By the end of CORTEX's engagement, individual OSS processes like resource design, reservations, configuration design, and provisioning were fully automated. The legacy BSS stack was replaced with a strategic one.
CORTEX was put head-to-head with existing RPA technology, but it was quickly recognised that parallel processing capabilities of CORTEX, plus the ability to re-use existing CORTEX architecture deployed in earlier projects not only showed CORTEX to be more performant, but also more cost effective.
A 99% service improvement was achieved.
The new digital product and services are now allocated within 20 seconds of ordering, while the previous variants of the product took more than 10 days for delivery.