Skip to main content

How to Efficiently Update Your Network Inventory System with Information from Technical As-built Documentation with the Help of AI?

GIS systems and digital models are crucial to the smooth operation of companies that manage networks, such as telecommunications, district heating, and energy. However, the systems alone are not enough—up-to-date and complete data is necessary for their smooth operation.

What will you learn from this article?

  • What are digital twins, and what are they used for?
  • Why is updating the digital network model quickly so important, and what are the challenges of manual digitization?
  • How to efficiently update a digital network model with the help of artificial intelligence and the iDoc solution?

What is a digital network model, and what is it used for?

Digital network models, also known as digital twins, are virtual representations of the physical network – electric, telephone, district heating, water supply, and others.

They allow tracking all the processes running on these networks and conducting analyses. They support both operational and strategic activities (planning modernization works, extensions, etc.). In general, they improve network management and, from a broader perspective, enterprise management.

Why is it so important to keep the data in a digital network updated?

For a digital network model to truly serve its purpose, it must be up to date – which actually means that the data it uses must be up to date. This will ensure that the model reflects the actual state of the network at any given time (and is indeed a “digital twin”).

Working with outdated data means working with a network that no longer exists, as seen in the GIS system. This means that it is impossible to effectively plan and execute tasks based on such a model, as any plans created and decisions made do not match the reality of the existing physical network.

This is why efficient and fast data updating in databases is so important. It allows you to keep up to date with all the processes on the network, react, and conduct analyses. This, in turn, translates into a significant reduction in the time it takes employees to perform tasks and ultimately into financial savings.

The investment process vs. updating the digital network model

The investment process in all companies with network infrastructure is very similar.

After the investment is carried out in the field, based on the technical design, construction works are accepted.

The investment’s as-built documentation is prepared. It is then archived in digital (scanned folders of documents) or paper form.

If the company uses a network inventory system (GIS), the information from the documentation is uploaded to it. This way, the digital model of the network is updated with the latest investments.

If the process of entering network information into the GIS system is carried out manually, it consists of the following steps:

  • Uploading the data about the designed network into the GIS system.
  • Completing construction and gathering the documentation
  • Verifying the document dossier for completeness of documentation
  • Scanning of documents
  • Entering facility geometry (spatial representation) information and other relevant data manually into the GIS database
  • Placing scans of documentation in the archive.

What problems are associated with manually entering network facility data into a GIS database?

While the process described above is accurate, entering information manually from as-built documentation takes a lot of time (and can result in data errors).

It can take several weeks or months from the time the project is completed to the time the information is digitized – that is, until the network model is updated.

In practice, this means that information about the latest investment, until it’s been entered into the network inventory system, is not involved in monitoring and troubleshooting outages, in processes of distribution (switching), billing, or analysis, and is not included in reports and fixed assets.

In the field, the network is already in place, while in the GIS system, it is still shown in the design phase. This can be compared to using an outdated road map. Planned roads or road turnouts are marked on it, but the GPS system does not consider them when planning the route, even if the road has already been built and could be freely used. What’s more, it would make getting to your destination much faster.

How can the problem of updating the digital network model be solved?

Artificial intelligence (AI) and the iDoc solution, developed by Globema, can come to the rescue.

The system uses artificial intelligence algorithms to read information from scans of technical documentation.

It categorizes the documents and reads the data they comprise. Then, the data is transferred to the GIS system. Thus, the network model is efficiently updated with information on the latest completed investments or upgrades.

The scanned documentation from which iDoc reads information is transferred to the digital document repository.

iDoc reads information from document pages, maps, schematics, and final reports – more than 30 attributes (including file number, contractor, facility, station, etc.) – and converts them into digital information. The number of attributes read by iDoc can be increased if needed.

What does documentation processing with the help of iDoc look like?

We load the scanned as-built documentation into iDoc.

The system verifies the completeness of the documentation dossier, then categorizes the documents, reads the objects and attributes, and provides suggestions for information to be input into the database (candidates).

The database candidates are ultimately verified by a human to ensure the highest accuracy of the read data and eliminate errors.

What are the benefits of digitizing as-built documentation with iDoc?

The process of automatically categorizing documentation, reading information from as-built documentation, and then transferring it to a database with the help of iDoc is about 15 times faster than manually entering information into a GIS system.

The iDoc system is also highly efficient at reading data. It identifies where data is missing and where verification is needed, eliminating typical human mistakes.

This affects both the work of the people involved in the digitization process and the ability to use the digital network model and network data:

Automatic digitization saves the time of persons dealing with manual information entry into the system. This allows them to take care of other tasks. Yet, they are still needed in the digitization process – to verify the correctness and approve the final version of the data.

Rapid digitization means that data can be updated faster in the network inventory system, and the operator can immediately work on an updated network model.

We eliminate the situation where the latest network information is not taken into account in decision-making processes (at the operational and strategic level), and as a result, the digital model does not fulfill the task for which it was created.

Automatic digitization guarantees high data quality. It allows to avoid errors that usually occur when information is manually transcribed into the system, the so-called human errors, resulting from tiredness from repetitive work, general fatigue, or distraction. The person responsible for verifying and accepting the final data will guarantee 100% correctness of the data (thanks to the step of approving or correcting the information sourced by iDoc).

iDoc reduces costs associated with manual documentation processing. This is due to better use of staff time and the lack of the need to hire people just to enter data into the GIS system.

Above all, however, iDoc helps reduce the financial losses resulting from working on an outdated data model that does not take into account the latest information on completed investments.

Did you find this article interesting? Are the challenges of efficiently digitizing network data and updating the digital network model also your concern?

Would you like to learn more about iDoc?  Visit the solution website or contact us directly: