Condition assessment of sewers and pipes: from CCTV inspection to surveying and AI

The results illustrate the scale of the task: around 18 per cent of the sewers surveyed were found to require renovation in the short to medium term. At the same time, the annual renovation rate stood at around 1 per cent of the public sewer network. Given a total length of just under 600,000 km, this corresponds to around 6,000 km per year.

Condition assessment is therefore far more than a technical inspection task. It provides the data basis for maintenance and investment planning. Network operators must be able to identify where action is required in the short term, which areas of damage should be monitored, and where rehabilitation measures can be deferred.

Recording, classifying and assessing

In practice, condition recording, condition classification and condition assessment are often grouped under the term ‘sewer inspection’. However, for systematic sewer management, a distinction must be made between the individual steps.

The first step is condition assessment. This involves recording the existing and discernible characteristics of a sewer. During visual inspection, this is carried out using camera systems. For example, cracks, pipe breaks, misaligned joints, corrosion, root ingress, deposits, protruding connections or deformations are documented.

This is followed by a standardised condition description. DIN EN 13508-2 and DWA-M 149-2 form the essential basis for this. The subsequent classification and assessment of the structural and operational condition is described in particular in DWA-M 149-3. The technical guideline was editorially revised in 2024. The aim is to present the need for rehabilitation, taking priority rankings into account.

It should be noted that whilst a camera image can document visible damage, it is not always possible to deduce from this alone what consequences this may have for structural stability, watertightness, operation or remaining service life. For renovation decisions, therefore, further factors such as age, material, hydraulic conditions, groundwater conditions, the progression of damage and the importance of the network section must be taken into account.

CCTV inspection remains the standard method

CCTV inspection remains the primary method for inspecting sewers that cannot be entered. Trolley systems fitted with rotating and swivelling cameras allow for a detailed examination of pipe walls, sockets and connections. Depending on the task at hand, satellite systems may also be used, enabling the inspection of branch pipes from the main sewer.

The quality of the subsequent assessment is determined during the inspection run itself. A sewer that has not been adequately cleaned may conceal damage, whilst travelling too quickly can cause details to be overlooked. Lighting, camera position and the careful examination of areas of concern also influence the result.

Added to this is the quality of the master data. A technically flawless recording loses considerable value if it is assigned to the wrong location or connection. Qualification and quality assurance therefore play a central role. The RAL-GZ 961 quality assurance standard for sewer construction explicitly covers, under Assessment Group I, the inspection of sewage pipes and sewers, and sets out corresponding requirements regarding specialist personnel, experience and technical equipment.

More information from a pipe inspection

Camera technology has advanced considerably in recent years. Higher resolutions enable more detailed images, whilst 360° systems can capture the interior surface of the pipe almost in its entirety. This allows the inspection and analysis to be separated more clearly: the inspection first generates a comprehensive set of image data, which can then be analysed in detail at the workplace.

This offers particular advantages in areas that are difficult to access. Areas of concern can be re-examined without necessarily requiring a second inspection run. At the same time, capturing as complete a picture as possible improves the conditions for automated analysis.

The inspection of service pipes is also evolving. Satellite technology not only enables their visual inspection; when combined with suitable sensors, it can in some cases also determine the spatial route of the pipe. This helps to fill gaps in older or incomplete inventory records.

Measuring rather than estimating

Alongside improved image quality, the quantitative recording of damage and geometric changes is becoming increasingly important. Deformations, pipe cross-sections, misalignments or positional deviations can be determined more objectively using suitable measurement methods than through a purely visual assessment.

Laser and profile measurements, for example, are used. These enable the geometry of a sewer to be recorded and deviations from the target cross-section to be determined. The combination of CCTV imagery and geometric measurement thus expands the data basis for assessment.

Particularly in the case of larger profiles or structural issues, visual inspection alone is often insufficient. Depending on the nature of the damage, additional investigations of the pipe-soil system, material tests or geophysical methods may be required. Visual inspection then forms one component of a more comprehensive structural assessment.

New perspectives through drones and mobile systems

The way in which cameras are moved through sewers and structures is also changing. In large or hard-to-access sections, drones can complement traditional inspection methods such as sewer inspections or on-site surveys. Using camera technology – and in some cases LiDAR – it is possible to survey areas that are difficult to reach with conventional systems.

In addition, manhole zoom cameras are used. These allow for a quick inspection of sewer sections from within the manhole and can be used, for example, for preliminary surveys or to shortlist areas for further investigation.

As a result, there is increasingly no single universal inspection method. Instead, operators have a toolbox of different systems at their disposal. Which method is suitable depends on the nominal diameter, accessibility, the specific issue, the risk of damage and the required quality of information.

AI supports the evaluation

The automated analysis of inspection data is currently developing at a particularly rapid pace. Given the large volumes of video and image material, the use of artificial intelligence is a natural choice. Algorithms can search through image data, identify conspicuous features and generate suggestions for damage descriptions or coding.

Real-world applications are already demonstrating that this allows parts of the condition assessment process to be automated. AI takes over recurring evaluation steps in particular and can help process large data sets more quickly and consistently.

However, expert oversight remains crucial. Rare damage patterns, poor recording conditions or complex situations can lead to misinterpretations. Accordingly, current systems often rely on a combination of automated preliminary analysis and subsequent review by qualified specialist staff.

Another key prerequisite is the quality of the input data. AI can only reliably analyse those features that are actually discernible in the source material. As a result, data cleaning, image quality and comprehensive data capture are becoming more important rather than being replaced by automation.

From inspection to sewer management

Digitalisation is ultimately changing the role of condition data. Inspection videos and damage codes no longer serve merely as evidence of a single survey. They are incorporated into sewer registers, GIS, rehabilitation planning and asset management systems.

If inspections are carried out over extended periods using comparable criteria, it is also possible to track the progression of individual defects. Consequently, alongside the current condition class, the focus is increasingly shifting to how rapidly a defect is changing. Such information can help to determine inspection intervals and rehabilitation timings in a more risk-based manner.

Technical progress therefore does not necessarily mean that traditional CCTV inspection will be replaced. Rather, it is becoming part of a more comprehensive digital process chain. Camera technology, 360° imaging, geometric surveying, drones and AI provide different types of information that can be combined depending on the specific requirements.

For network operators, the key benefit does not lie in acquiring as much data as possible. What is crucial is reliable, reproducible and, in the long term, comparable information. It is only through a qualified condition assessment, a traceable classification and a professional evaluation that the foundation for an economically and technically sound rehabilitation strategy is established.

Sources

German Association for Water, Wastewater and Waste (DWA): Condition of the sewerage system in Germany, eighth DWA sewerage survey, published in 2020, based on 2018 data.

DWA-M 149-2: Condition assessment and evaluation of drainage systems outside buildings – Coding system for visual inspection.

DWA-M 149-3: Condition assessment and evaluation of drainage systems outside buildings – Part 3: Evaluation following visual inspection, 2024 edition.

Quality Assurance in Sewer Construction: Quality and Testing Specifications RAL-GZ 961, Assessment Group Inspection (I).

3R 01–02/2024: Smart Monitoring with Artificial Intelligence (AI).

3R 04–05/2021: Sewer cross-section and profile surveying in practice.

3R 07–08/2020: From condition assessment to implementation – database-driven working as a strategy.

3R 04–05/2017: Special methods for assessing the condition of sewer systems.

3R 07–08/2016: Technical innovations in the condition assessment of sewer systems.

3R, 2026: From drones to AI: Smart technologies for sewer operations.

Manufacturer information from IBAK, JT-elektronik and CanSol on TV inspection, 360° camera technology, surveying methods and AI-supported analysis.