PipeAId is a computer vision AI system that incorporates machine learning algorithms that capture and recreate the workflow and processes used to document observations for sewer inspection.
Think of PipeAId as a person and the computer vision component as the eyes that examine the CCTV video and extract data such as taps, cracks, roots and other obstructions. The machine learning component is the brain that gets smarter and more accurate each time it processes more data. When a defect is present, the AI determines which National Association of Sewer Service Companies (NASSCO) Pipeline Assessment Certification Program (PACP) code applies.
PipeAId’s streamlined approach takes less time than a traditional sewer assessment. With our PipeAId data collection methods, contractors can collect more sewer video each day in the field because they no longer need to code the defects on-site. It is also designed to work with standard sewer inspection equipment and operating procedures, which means there is no need for expensive specialty software or retraining of staff on how to inspect pipes. PipeAId lowers costs, as demonstrated recently in a project in an Ohio community where sewer inspection costs were reduced by 50% with PipeAId over traditional methods.
Many factors can impact how a technician codes a pipe segment. PipeAId removes subjectivity to ensure that observations are coded correctly. Ensuring that the codes are accurate is paramount, as the data is used by utility owners to make decisions about their systems. Machine learning continually increases PipeAId’s intelligence and defect recognition. As more data is processed, images are fed into the AI models to continuously train its computer vision. This continuous training ensures the AI understands every aspect of the defect and the different ways the defect can present itself in a collection system. The more PipeAId is applied, the more the confidence interval of the coding output improves. It also allows the technician to focus on cleaning the pipe and leave the coding to PipeAId.
To provide a comprehensive overview of your pipelines, PipeAId uses geographic information system (GIS) output with a digital twin of the collection system that geospatially locates each defect and observation. The digital twin included in the GIS database that customers receive shows exactly where each root, tap, break, sag, etc. is located. The database also documents pipe data such as materials, size, upstream/downstream manholes, NASSCO PACP scoring and each defect/observation. Paired with a data hosting solution, the digital twin links the inspection video to each pipe segment so that staff can access it in the field or in the office by simply clicking on the line segment. This high level of detail gives a complete picture to field staff, contractors and decision-makers as they determine when and where to maintain, rehabilitate, repair or replace pipes in their system.
PipeAId is designed to be easily integrated into a community’s existing workflow by leveraging current standard operating procedures and existing equipment/software. It does not need other components that legacy software requires, such as a database. Results are open source and can be leveraged through GIS and CADD software that most communities already have. PipeAId does not hold data hostage but, rather, empowers communities to be more efficient in their operations while saving money.
Reach out to us today for a free demonstration so you can see how PipeAId can benefit your town or utility.