How to do a TMH 9 visual assessment on a flexible road pavement

A TMH 9 visual assessment is a pavement condition survey in which an accredited assessor rates every pre-defined road segment for each distress by type, degree (severity) and extent (how widespread along the segment), on a standard form. This guide sets out the order of work for a flexible pavement under Parts A and B of TMH 9, the South African Manual for Visual Assessment of Road Pavements (Committee Draft Final, May 2016) of COTO, the Committee of Transport Officials, published by the South African National Roads Agency. (COTO, 2016b) (COTO, 2016j)

Key facts

• TMH 9 Part A (Committee Draft Final, May 2016) records each pavement distress by type, degree and extent, and Part B supplies the flexible-pavement descriptions and the assessment form.
• TMH 9 Part A, section A.3, recommends that the road authority predefines assessment segments across its network instead of leaving them to individual assessors.
• TMH 9 Part A, section A.2.4, says that where a pavement distress shows several degree and extent combinations, the combination with the highest product of the two is recorded.
• TMH 9 Part A, section A.5.1, limits visual assessment to dry conditions and tells the assessor to concentrate on the worst lane of a multi-lane road.
• TMH 9 Part B, section B.5, says the overall pavement condition on the form is not used in data processing and serves as a check on the other ratings.
• Draft TMH 22 (March 2013), the South African road asset management manual, gives an example routine in Appendix J-4 that checks TMH 9 ratings for conflicting combinations.

Which parts of TMH 9 does a flexible pavement assessment use?

A flexible-pavement assessment opens Part A first, then rates each segment against Part B, which refers back to Part A for degree and extent (COTO, 2016j). Part A is general and holds the procedure, quality management plan and safety provisions. Part B covers flexible pavements; Parts C, D and E cover concrete, block and unpaved roads (COTO, 2016b). TMH 9 belongs to the Technical Methods for Highways (TMH), the COTO series of South African road manuals.

Part A states that it replaces the 1992 draft, and its version note says committee drafts have no legal standing. The client's terms of reference decide which edition governs a programme: name the edition and form on the first page of each report. TMH 9 is a companion to Draft TMH 22 (March 2013), the road asset management manual (COTO, 2016b) (COTO, 2013).

How do you do a TMH 9 visual assessment, step by step?

Thirteen steps run from the authority's programme to the hand-over file, each with its clause and the record to keep.

Narrated film · 1:21

Running a TMH 9 visual assessment

A step rail along a synthetic road beside one assessment segment that gains rated distresses and a photograph reference.

Read the film transcript

How do you run a TMH 9 visual assessment, a South African pavement condition survey, so every rating can be reviewed later? A segment is the length of road that receives one rating. Each distress, a visible defect, is recorded by type, by degree, its severity, and by extent, how widespread it is. First fix the segment boundaries, appoint accredited assessors, and plan the survey window and the hazard assessment. Record the inventory, choose the worst lane of a multi-lane road, then rate the surfacing, the structural distress and the functional items. Complete the form, then photograph what it cannot carry, a step we recommend because neither part asks for photographs. Before the crew leaves, verify the ratings for conflicting pairs, then hand them over with the quality record and the edition used. Where a distress shows several combinations, record the highest product of degree and extent, not simply the worst degree. In this synthetic segment, cracks are recorded at degree three, extent two, and rutting at degree two, extent three. The one deep rut is absent from the recorded rating, so the comments carry its chainage and a photograph reference. Keep the degree, the extent and the photograph together, so a reviewer can see what the assessor saw. TMH 9 stops at ratings, so next read Draft TMH 22 for condition indices and TRH 12 for investigating a flagged road.

1. Fix the segment boundaries. A segment is the length of road that receives one assessment rating (A.7). Part A recommends that the road authority predefines segments over its network instead of leaving the choice to assessors; Table A.3 gives standard lengths, and segments are not uniform sections (A.3) (COTO, 2016b). Keep: the segment list, with start and end chainage, the distance along the road from its start.
2. Appoint accredited assessors and calibrate. Only assessors with a valid accreditation for the pavement type carry out assessments (A.5.3). Internal quality control begins with a calibration session on network-specific issues (A.5.4) (COTO, 2016b). Keep: the assessor register and the calibration notes.
3. Set the survey window. Part A prefers assessment in or towards the end of the rainy season and in the cooler months, within a limited period, and allows dry conditions only, because a wet road hides distress. It also limits average daily length and driving speed; take the figures from A.5.1 (COTO, 2016b). Keep: the planned dates and each survey day's weather.
4. Complete the hazard assessment. Before mobilising, carry out a hazard identification and risk assessment (HIRA) for road inspections (A.6). Part A's equipment includes a flashing warning light, signs, the authority's permission letter and high-visibility clothing (Table A.6) (COTO, 2016b). Keep: the signed HIRA and the permission letter.
5. Record the segment inventory. Table A.4 marks five inventory items as required: road number or name, section or link, segment start and end, assessor name and date. Each optional item distracts the assessor from the condition rating (A.4) (COTO, 2016b). Keep: the form header, with segment length and width in metres (B.6) (COTO, 2016j).
6. Orient on the first segment and choose the lane. Sun position, cloud cover and local moisture change how visible cracks are, so the first segment needs a thorough orientation. Assess the full paved width, including paved shoulders, and on a multi-lane road the worst lane, typically the slow lane (A.5.1) (COTO, 2016b). Keep: the lane and direction assessed, and why any segment was not assessed.
7. Rate the surfacing. A distress, a visible defect in the pavement, is recorded by type, degree (its severity) and extent (how widespread along the segment) (A.2). Part B.2 describes seven surfacing distresses, including cracks, aggregate loss and bleeding; the form adds texture and voids. Dot the form as each is seen, then mark one degree and extent per distress at the segment end (A.5.1). Where one distress shows several degree and extent combinations, record the one with "the highest product of the two" (A.2.4), a rule it calls generic, varying by distress type and assessment level (COTO, 2016b) (COTO, 2016j). Keep: the surfacing block of the form.
8. Rate the structural distress. Part B.3 describes cracking (block, longitudinal, transverse and crocodile), pumping, deformation (rutting and undulation), structural patching, and failures or potholes. Rate each by degree and extent under the same rule. Sealed cracks are not rated unless they have reopened (B.3.1) (COTO, 2016j). Keep: the structural block of the form.
9. Rate the functional items. Part B.4 covers roughness, skid resistance, surface drainage, unpaved shoulders and edge defects. The form marks a degree for each, an extent too for edge defects, and the problem behind it. Side-drain problems are noted, not rated (B.4.3.2) (COTO, 2016j). Keep: the functional block of the form.
10. Complete the summary and the form. Part B.5 records the overall pavement condition as one degree, plus other problems such as service crossings. It is not used in data processing; it checks the other ratings (COTO, 2016j). Re-check segment start, finish and width (A.5.1), then add the assessor's name and the date (B.6) (COTO, 2016b). Keep: the completed form.
11. Photograph what the form cannot carry. Neither part asks for a photograph; this step is this guide's recommendation, not a clause. Take one per segment, and one of any localised occurrence the highest-product rule leaves out of the rating, named by road, direction and chainage. Keep: the photographs, referenced in the form's comments.
12. Verify before the crew leaves the area. Draft TMH 22 gives severe bleeding beside very good skid resistance as a conflicting pair (D.5.11), and its Appendix J-4 is an example routine that pairs ratings and attaches an error message; the limits are in the appendix (COTO, 2013). Also carry out the follow-up check A.5.4 sets for the first days (COTO, 2016b). Keep: the flagged segments and their outcomes.
13. Hand over the ratings with the quality record. Quality acceptance is independent re-assessment of a representative sample of roads by experienced accredited assessors; A.5.4 sets the share and time limit (COTO, 2016b). Hand the ratings over in the format the authority names, with its version, for example the TMH 18 road-data file formats (Draft Version 4, October 2014; Committee Draft 5.0, February 2016) (COTO, 2014) (COTO, 2016i). Keep: the file as sent, the quality record and the editions used.

What does one synthetic segment look like once it is rated?

Synthetic example: every value below is invented. Segment S1-07 of synthetic two-lane road S1 runs from km 14.20 to km 16.10 (1,900 m long) and is 6.8 m wide. The assessor rates the full paved width, driving in the direction of increasing chainage on a dry day.

Distress	Observed	Degree	Extent	Product
Transverse cracks	Faint, widespread	1	4	4
Transverse cracks	Opened, few places	3	2	6
Rutting	Shallow, intermittent	2	3	6
Rutting	Deep, one place (km 15.35)	5	1	5
Failures / potholes	One large, deep pothole	5	1	5

Under A.2.4, record transverse cracks at degree 3 and extent 2, rutting at degree 2 and extent 3, and the pothole at degree 5 and extent 1, its only combination (COTO, 2016b). Check value: each recorded combination has the highest product among the rows for its own distress.

The deep rutting at km 15.35 is absent from the recorded rating. The A.2.4 examples treat a severe isolated occurrence as localised distress requiring routine attention, so write it in the form's comments with its chainage and a photograph reference, for example S1-07-04 (COTO, 2016b). The rating tables stay in Parts A and B; nothing here converts the codes to an index. The insight on degree and extent as evidence covers the arithmetic of extent.

What does TMH 9 leave to the road authority?

TMH 9 supplies the procedure and the ratings. Three decisions sit outside it.

• The index. Part A says visual assessments determine visual condition indices (VCIs) for road pavements (A.1.1), but Parts A and B give no formula. Draft TMH 22 describes one variant in Appendix J-6 (E.2.1), so name the variant in every report; the guide on calculating a VCI from TMH 9 ratings and the insight on VCI mathematics take this step (COTO, 2016b) (COTO, 2013).
• The treatment. Part B.3 says the assessments are used to determine the need for maintenance, but Parts A and B hold no trigger rules for that step. Draft TMH 22 Part G describes needs determination, and Appendix J-8 gives example trigger algorithms (COTO, 2016j) (COTO, 2013).
• The level of detail. Network and project assessments use one method, and project level needs more detail on extent and location (A.2.1) (COTO, 2016b).

Which misreadings of TMH 9 should you avoid?

• Recording the worst degree. The rule does not simply select the most severe degree. In the A.2.4 examples a lower degree with a wider extent is recorded because its product is higher, and the severe isolated occurrence is treated as localised distress (COTO, 2016b).
• Overestimating extent. Part A notes that even experienced assessors tend to overestimate extent, and more so as severity grows (A.2.3). Judge extent against the whole segment, not the worst stretch (COTO, 2016b).
• Treating the overall condition as an input. The overall degree in B.5 checks the other ratings; it does not feed data processing (COTO, 2016j).

What does the hand-over file hold?

The authority names the format; where it names TMH 18, the version matters because the drafts differ. TMH 18 (Draft Version 4) section 5.2 describes the flexible-pavement visual condition summary as a .vcf file: road identifier, lane code, start and end kilometre and measurement date, then index fields and a degree and an extent for each surfacing and structural distress (COTO, 2014). TMH 18 (Committee Draft 5.0) describes it in section 5.2 as a .flx file (Table 5), adds an authority identifier and start coordinates, and makes the lane code optional unless the data describe one lane (COTO, 2016i).

Both versions list index fields, including a VCI, beside the ratings, so agree with the receiver which VCI variant, if any, the file carries. Both discard values outside the road's start and end kilometre, so check each segment against the network file. The quality control map shows where these checks sit.

What should you keep after the assessment?

For a reviewer a year later, keep:

• the segment list and the register extract behind it;
• the assessor register and calibration notes;
• each completed form, naming the edition;
• the photographs, filed by segment and chainage;
• the verification output, the A.5.4 quality record and the hand-over file as sent.

The procedure, step by step

1. Fix the segment boundaries

   Use the segments the road authority predefines, as Part A recommends (A.3); Table A.3 gives standard lengths. A segment receives one rating (A.7). Keep: the segment list, with start and end chainage, the distance along the road from its start.

2. Appoint accredited assessors and calibrate

   Appoint only assessors with a valid accreditation for the pavement type (A.5.3), and begin internal quality control with a calibration session on network-specific issues (A.5.4). Keep: the assessor register and the calibration notes.

3. Set the survey window

   Assess preferably in or towards the end of the rainy season and in cooler months, within a limited period. Assess in dry conditions only, within A.5.1's limits on average daily length and driving speed. Keep: planned dates and daily weather.

4. Complete the hazard assessment

   Before mobilising, carry out a hazard identification and risk assessment (HIRA) for road inspections (A.6); Part A's equipment includes a flashing warning light, signs, the authority's permission letter and high-visibility clothing (Table A.6). Keep: the signed HIRA and permission letter.

5. Record the segment inventory

   Record the five required inventory items (Table A.4): road number or name, section or link, segment start and end, assessor name and date; optional items distract the assessor (A.4). Keep: the form header, segment length and width in metres (B.6).

6. Orient on the first segment and choose the lane

   Orient thoroughly on the first segment, then assess the full paved width, including paved shoulders, and on a multi-lane road the worst lane, typically the slow lane (A.5.1). Keep: lane and direction assessed, and why any segment was not assessed.

7. Rate the surfacing

   Rate surfacing distresses (B.2) by degree (severity) and extent (how widespread) (A.2). Where a distress shows several combinations, record the one with "the highest product of the two" (A.2.4), a rule the manual calls generic. Keep: the surfacing block.

8. Rate the structural distress

   Rate each structural distress in Part B.3 (cracking, pumping, deformation, structural patching, failures or potholes) by degree and extent, using the A.2.4 rule; do not rate sealed cracks unless reopened (B.3.1). Keep: the structural block of the form.

9. Rate the functional items

   Rate roughness, skid resistance, surface drainage, unpaved shoulders and edge defects (B.4): a degree for each, an extent too for edge defects, and the problem behind it. Note side-drain problems without rating them (B.4.3.2). Keep: functional block of the form.

10. Complete the summary and the form

    Record the overall pavement condition as one degree, plus other problems (B.5): a check on the other ratings, not used in data processing. Re-check segment start, finish and width (A.5.1); add assessor name and date (B.6). Keep: the completed form.

11. Photograph what the form cannot carry

    Photograph each segment and any localised occurrence the highest-product rule leaves out of the rating, named by road, direction and chainage; neither part asks for photographs, so this is the guide's recommendation. Keep: the photographs, referenced in the form's comments.

12. Verify before the crew leaves the area

    Check ratings for conflicting pairs: Draft TMH 22 gives severe bleeding beside very good skid resistance as a conflicting pair (D.5.11) and an example check routine in Appendix J-4. Run the A.5.4 follow-up check. Keep: flagged segments and outcomes.

13. Hand over the ratings with the quality record

    Hand over the ratings in the authority's named format and version, for example TMH 18 (Draft Version 4, October 2014; Committee Draft 5.0, February 2016), with the quality record (A.5.4). Keep: the file as sent, quality record and editions used.

Frequently asked

Which edition of TMH 9 applies to a visual assessment?

The South African road manual TMH 9 (also written TMH9) of COTO, the Committee of Transport Officials, exists as a Committee Draft Final dated May 2016 for all five parts. Part A replaces the 1992 draft, Part E withdraws Draft TMH 12, and Part A says committee drafts have no legal standing. The client's terms of reference therefore decide which edition governs a programme, so name the edition and form on the report's first page.

How long is a TMH 9 assessment segment?

The South African road manual TMH 9, Part A, Table A.3, gives the standard assessment length for each road type, with a separate rule for urban blocks and a note limiting how far a segment may depart from the standard. Part A also recommends that the road authority predefines the segments across its network. Read the lengths in the table itself; RoadSense does not reprint them.

When should a TMH 9 visual assessment be carried out?

Section A.5.1 of Part A of the South African road manual TMH 9 prefers visual assessment in or towards the end of the rainy season and in the cooler months, completed within a limited period. It limits assessment to dry conditions, because a wet road hides distress and leads to erroneous ratings. The same section limits the average daily length surveyed and the driving speed; read the figures there.

Who may carry out a TMH 9 visual assessment?

Section A.5.3 of Part A of the South African road manual TMH 9 describes a national accreditation of visual assessors, held for each pavement type, and says only assessors with a valid accreditation may carry out assessments. First-time candidates are accredited initially as assistant assessors. Section A.5.2 describes the training that precedes accreditation; the tests, sample sizes and validity period are set out in those sections and are not repeated here.

Does a TMH 9 visual assessment need photographs?

Parts A and B of the South African road manual TMH 9 do not ask for a photograph for each segment; the Part B assessment form records degree and extent, with a comments box. A photograph is still good practice, because it lets a reviewer see what the assessor saw. Name each photograph by road, direction and chainage, the distance along the road from its start, and reference it in the form's comments.

Where can the official TMH 9 files be downloaded?

The Department of Transport publishes the five parts of the South African road manual TMH 9 as PDF files on its roads engineering standards and manuals page, and RoadSense does not host them. Download the part you need and read the cover to confirm the edition, which for the 2016 files is Committee Draft Final, May 2016. Part A and Part B are the two parts a flexible-pavement assessment uses.

Where this connects

Draft TMH 22 (March 2013) Part E, section E.2.1, describes the condition indices computed from these ratings, and from a visual rating to a rehabilitation shortlist follows one rating through that chain (COTO, 2013). TRH 12 (Technical Recommendations for Highways 12, Flexible Pavement Rehabilitation Investigation and Design, Draft, 1997) sets out a systematic procedure of investigation, evaluation and analysis of the existing pavement, then rehabilitation design and economic appraisal, for a flagged road (Department of Transport, 1997).

Last reviewed 21 October 2026 against the editions named above.

References

1. COTO (2016b). TMH9: Manual for Visual Assessment of Road Pavements. Part A: General. Committee Draft Final, May 2016. Original source.Section A.2: degree and extent
2. COTO (2016j). TMH 9: Manual for Visual Assessment of Road Pavements. Part B: Flexible Pavements. Committee Draft Final, May 2016. Original source.Sections, tables and clauses are cited inline in each guide
3. COTO (2013). Draft TMH 22: Road Asset Management Manual. Draft, March 2013. Original source.Sections, tables and clauses are cited inline in each guide
4. COTO (2014). TMH 18: Road Asset Data Electronic Exchange Formats. Draft Version 4, October 2014. Original source.Sections, tables and clauses are cited inline in each guide
5. COTO (2016i). TMH 18: Road Asset Data Electronic Exchange Formats. Committee Draft 5.0, February 2016. Original source.Sections, tables and clauses are cited inline in each guide
6. Department of Transport (1997). TRH12: Flexible Pavement Rehabilitation Investigation and Design. Draft TRH12, Pretoria: Department of Transport for COLTO. Original source.Chapter 1: rehabilitation design process; investigation, evaluation, analysis and economic appraisal

Use the manual edition and project requirements that apply to your work. Worked examples use synthetic values and are not site-specific design advice.