Network

Network May 2019

Issue link: https://fhpublishing.uberflip.com/i/1111696

Contents of this Issue

Navigation

Page 14 of 47

NETWORK / 15 / MAY 2019 INDUSTRY INSIGHT The future of utility quality assurance With the growing investment in domestic and commercial developments alongside the daily replacement and repair of our network assets, quality installation procedures and conformance to industry standards has never been more important for the future performance of our infrastructure. Mark Pereira, marketing manager at ControlPoint, discusses. T o ensure we create networks that last for their expected design-life, the industry must continue to innovate and focus on systems that make quality assurance robust and consist- ent. Over the past few years, there have been signi cant developments in quality assur- ance systems, ranging from the capture, monitoring and storage of installation data, to how we train and maintain the skills of those on the ground, installing our network assets. It is important to maintain a holistic approach to quality systems to ensure their overall e• ectiveness. For example, basic methods of capturing data from on-site equipment provide records of assets but not of their quality and it is important to recognise that if we install assets that do not conform to industry standards, then simply capturing data has limited value. Further to this, we are potentially accruing signi cant health and safety risks. This is one of the reasons that not only capturing but inspecting data from site is now widely regarded within the industry as an integral part of the asset installation process. Where remote, third-party inspections have taken place, the quality and longevity of networks have increased to levels previously regarded as out of reach. From the early days of simple photo- graphic data capture and information logging, to the now widely-utilised BlueBox data capture system and remote inspection services, ControlPoint has supported the growth and development of our networks by providing cross-utility quality assurance. With data and insight gathered from the installation of millions of network assets, ControlPoint is now uniquely positioned to develop the next level of quality assurance with the application of arti cial intelligence and deep learning. Using arti cial intelligence and machine learning, ControlPoint is developing the ro- bustness and consistency of remote inspec- tion by training its systems to recognise key quality criteria based on network industry standards. The AI system has been trained to recognise non-conformances within the electrofusion joint assembly process such as un-clamped assemblies, insuƒ cient scraping and contamination on and around the joint. How the system works 1. On-site operatives use the ControlPoint BlueBox mobile application to record photographs of joint assemblies. 2. These photographs are instantly checked against key quality criteria by the AI system. 3. Automatic alerts of non-conformance are sent to the operative to rectify issues and assist in the creation of compliant assets. Although the system has been developed to assist with the installation of electro- fusion assets within the gas and water industries, its principles can be rolled out to support quality assurance across any number of applications. Alongside industry innovations such as this, we must not forget the important role of ongoing training and support in the com- missioning of safe and conformant assets and continuous improvement. All things considered, a combination of robust data capture, inspections and training systems holds the key to real quality assurance and long-term network performance.

Articles in this issue

Archives of this issue

view archives of Network - Network May 2019