Right-of-Way Maintenance Prioritization by Applying Machine Learning Models on Third-Party Call Notification Data
Proceedings Publication Date
Vinicius C. Peixoto
Vinicius C. Peixoto, Paulo F. N. Afonso
Part of the proceedings of

Public awareness programs are an important factor in preventing damage to pipelines. As part of a public awareness program, a pipeline operator’s call center provides community outreach, building trust and better relationships with the public along the right-of-way (ROW). Furthermore, an effective operator’s call center can reduce resistance to pipeline maintenance and ROW activities, preserve ROW, enhance emergency response coordination, and improve pipeline operator reputation. Ensuring proper response from operators to call notifications, including making right-of-way maintenance improvements as demanded by the community call notifications, is fundamental to keeping a high level of engagement of the affected public who live or work near a pipeline. In Brazil, this public is commonly the first stakeholder to report any suspected illegal or unauthorized third-party actions and ROW encroachment. Against this background, we wrangled almost 6,000 call notification data from 48 ROW segments from the Metropolitan Area of São Paulo City, Brazil. Unsupervised Machine Learning, such as principal component analysis (PCA) and its factors were used to build a rank of ROW segments based on the number of call notifications and their categories. The rank can be used to support a data-based decision for the prioritization of resources for ROW maintenance.

To view the video or download the paper please register here for free

You already have access? Sign in now.