What belongs in a road-condition survey acceptance pack?

Twelve thousand records, one awkward gap

A folder can look reassuringly full while the evidence register tells a different story.

TMH13 Part C includes calibration, validation and operational quality-control guidance for roughness measurements. The requirements depend on the measurement approach. (COTO, 2016a)

Reconcile twelve files before accepting seven

Governing relationship: Accepted / received = 7 / 12 = 58.3%.

NR = received records; NA = accepted records. Checked is an intermediate state.

Twelve received, nine checked, seven accepted. The sets overlap; do not add them.

The film first plots records, then follows the calculation and plots share of received (%). Seven records are accepted; five are unresolved. Of those five, three await checking and two were checked but not accepted.

Set accounting

Evidence and acceptance

A receipt count cannot substitute for an acceptance decision.

Download example data (CSV)

Model, calculations and transcriptEquations, worked steps, source notes and a timed reading guide

Understand the model

Accepted records are a subset of checked records, which are a subset of received records. A fraction is meaningful only when its denominator is clear.

Work through the inputs and equations below, then compare the calculated output with the film. All numerical inputs in this study are synthetic.

Inputs and symbol key

NR; NC; NA

Numbers of received, checked and accepted records.

fA

Accepted fraction of received records (dimensionless).

Values used in this film

• NR = received records; NA = accepted records. Checked is an intermediate state.
• Twelve received, nine checked, seven accepted. The sets overlap; do not add them.

The mathematical relationships

1. Equation 1

   Source and scope
   $$f_A=N_A/N_R$$

   Divide accepted records by received records for the receipt-based acceptance fraction.

2. Equation 2

   Source and scope
   $$7/12=0.5833;\quad N_R-N_A=5$$

   Seven accepted out of twelve leaves five unresolved: three unchecked and two checked but not accepted.

Work through the example

Use the relationships above with consistent units. These substitutions are our worked example, not values quoted from the source.

1. Unchecked

   12 − 9 = 3

   records

2. Checked, not accepted

   9 − 7 = 2

   records

3. Partition check

   3 + 2 + 7 = 12

   records

4. Acceptance fraction

   7 / 12 × 100 = 58.3333

   percent

Read the output

Seven records are accepted; five are unresolved. Of those five, three await checking and two were checked but not accepted.

Check your understanding

Reconcile all twelve identifiers to the three exclusive states. Reject a snapshot in which accepted exceeds checked or checked exceeds received. Verify that re-deliveries update lineage instead of silently duplicating accepted coverage.

Assumptions and limits

A receipt count cannot substitute for an acceptance decision.

Return to the full references and method notes to check the cited edition and its applicability.

Timestamped film transcript 27 short entries · select a time to revisit the film

1. 0:00

   Evidence and acceptance. Twelve files; nine checked; seven accepted.

2. 0:01

   MATHEMATICS · define the relationship and the meaning of each symbol before calculating.

3. 0:07

   NR = received records; NA = accepted records. Checked is an intermediate state. Twelve received, nine checked, seven accepted. The sets overlap; do not add them.

4. 0:14

   INPUT · nested record counts · read the axis units and the numerical values.

5. 0:23

   Twelve files; nine checked; seven accepted.

6. 0:25

   Received: 12 of 12 scheduled records (100.0%). Grey records remain visible.

7. 0:30

   Checked: 9 of 12 scheduled records (75.0%). Grey records remain visible.

8. 0:35

   Accepted: 7 of 12 scheduled records (58.3%). Grey records remain visible.

9. 0:43

   OUTPUT · accepted and unresolved shares · read the axis units and the numerical values.

10. 0:52

    Seven records are accepted; five are unresolved. Of those five, three await checking and two were checked but not accepted.

11. 0:59

    CALCULATION · substitute the defined inputs; keep units through each intermediate step.

12. 1:05

    Unchecked: 12 − 9 = 3 records

13. 1:08

    Checked, not accepted: 9 − 7 = 2 records

14. 1:11

    Partition check: 3 + 2 + 7 = 12 records

15. 1:14

    Acceptance fraction: 7 / 12 × 100 = 58.3333 percent

16. 1:17

    SENSITIVITY · change one declared input · read the axis units and the numerical values.

17. 1:26

    Current register: 58.3333 % accepted. Seven accepted among twelve received.

18. 1:29

    One resolved: 66.6667 % accepted. One existing unresolved record becomes accepted.

19. 1:33

    Two new unresolved: 50 % accepted. Two additional records arrive; accepted count stays seven.

20. 1:36

    Check: Reconcile all twelve identifiers to the three exclusive states. Reject a snapshot in which accepted exceeds checked or checked exceeds received. Verify that re-deliveries update lineage instead of silently duplicating accepted coverage.

21. 1:42

    ONE ROW · A folder can look reassuringly full while the evidence register tells a different story.

22. 1:55

    PARAMETER MAP · 12000 evaluated cells. The colour limits stay fixed across every row.

23. 2:05

    The 60 by 200 matrix contains 12,000 synthetic receipt records. Codes 0, 1 and 2 mean unchecked, checked but unresolved, and accepted. The modular rules deliberately place

24. 2:15

    reproducible gaps. They are data-generation rules, not an acceptance algorithm. Scan a row to find exceptions within a batch, then scan vertically to see repeated delivery

25. 2:24

    patterns.

26. 2:29

    INTERPRETATION · These are discrete states. Read the map cell by cell; a smooth surface would invent intermediate categories.

27. 2:38

    A receipt count cannot substitute for an acceptance decision.

Download the timed captions (VTT)

The transcript records the on-screen wording. Formula provenance and interpretation are explained in the source note above.

Build an evidence pack

Our suggested pack has an inventory, method/edition record, collection log, quality-control evidence, exception register and accepted outputs. Give each item a stable identifier. Keep a missing item visible; an empty cell should not quietly become a pass.

Review a missing calibration record

A synthetic delivery contains measurement files and a summary but lacks the referenced calibration report. Mark the evidence item as unresolved, assign an owner and record what is needed. File presence alone does not demonstrate measurement acceptance.

Make exceptions traceable

This editorial checklist is not a replacement for a contract specification. Agree the actual acceptance criteria with the responsible engineer before collection begins.

Use explicit evidence states

A useful acceptance register distinguishes received, checked, accepted and unresolved. These are suggested workflow states, not contract clauses. A file may be received while its contents are still unchecked; a checked file may contain an unresolved exception.

For each required item, record its identifier, version, responsible reviewer, decision date and the criterion used. Link a summary to the exact input version. If a corrected survey replaces an earlier delivery, identify which reports need to be regenerated.

A concise handover should let the recipient answer three questions: which outputs are being accepted, what evidence supports that decision, and what limitations remain? This is more useful than a folder containing many files with no record of their relationship.

Reconcile the handover before accepting the summary

Consider a synthetic delivery requiring 12 evidence items. All 12 have arrived, nine have been checked, and seven have been accepted. The correct conclusion is not “the delivery is complete”. It is that receipt is complete while acceptance remains incomplete. Five items remain unresolved: two have been checked without acceptance, and three have not yet been checked.

The acceptance fraction is 7 / 12 = 58.3%. The checked fraction is 9 / 12 = 75%. These describe different workflow states. Their denominators are the same agreed inventory, not the number of files a reviewer happened to open.

Case review

Twelve received. Seven accepted. What happened to five?

Which records still prevent an unambiguous handover?

Unchecked · 3 Records

The discriminating check

Inspect the acceptance register rather than using the existence of a file as proof of acceptance.

Evidence supports the comparison

Release the seven accepted records with their acceptance evidence and an explicit statement of coverage.

Evidence is missing or incompatible

Keep the other five visible as exceptions; request the missing checks or resolve the recorded findings.

What you can conclude. A 58.3% accepted fraction describes this delivery state. It is not a survey-accuracy score.

Original worked case. Read the source editions and method notes.

A received item may contain an incorrect identifier, an unsupported claim of calibration, or an unresolved exception. Conversely, an item awaiting review is not necessarily defective. An effective register keeps these situations separate so that the next action is clear.

Evidence state	Interpretation	Next action
Received, unchecked	Presence established; content not reviewed	Assign a reviewer
Checked, unresolved	A question or exception remains	Record the specific resolution needed
Accepted	Defined criterion met by an identified reviewer	Preserve the accepted version
Superseded	A later version replaces this record	Trace affected outputs

TMH13 Part C provides the measurement-quality context for roughness surveys. The register above is an original workflow example, not a replacement for its requirements or a contract's acceptance schedule. (COTO, 2016a)

A useful handover names the accepted outputs and the evidence version behind each. If a corrected measurement file arrives after acceptance, identify the summaries, charts and exports derived from the previous version. Reopening the relevant decisions is more defensible than replacing a file while leaving its downstream reports untouched.

The goal is not a ceremonial checklist. It is a record that lets the receiving engineer establish what can be relied upon, what remains provisional, and who is responsible for resolving the remaining questions.

Calculation laboratory

How do we reconcile the register without counting the same record twice?

Calculation step	Substitution or result	Units and interpretation
Unchecked	12 − 9 = 3	records
Checked, not accepted	9 − 7 = 2	records
Partition check	3 + 2 + 7 = 12	records
Acceptance fraction	7 / 12 × 100 = 58.3333	percent

A useful register has mutually exclusive terminal states for a particular review snapshot. Received and checked are cumulative membership states, so their bars are not additive. The partition into unchecked, checked-but-not-accepted and accepted is additive because each received record belongs to exactly one of those groups. This is a set-accounting calculation, not a measure of pavement condition.

The denominator can move independently of review quality. Resolving one existing record raises acceptance to 66.67%, whereas receiving two new unresolved records lowers it to 50% without reversing any earlier decision. A dashboard that reports only a percentage can therefore imply declining performance during successful data delivery. Show counts, receipt dates and decision dates with the fraction.

File-level acceptance also differs from kilometre coverage. A short file and a long file each contribute one record here. If the intended question concerns accepted survey distance, sum unique accepted intervals on the route reference and divide by unique required distance. Overlapping deliveries must not create extra surveyed kilometres. The calculation should follow the contractual acceptance unit and preserve reasons for rejection, rework or partial acceptance. TMH13 provides measurement context; this original register example does not introduce a contractual pass threshold. (COTO, 2016a)

Controlled sensitivity experiment

Controlled case	Calculated output	What changes
Current register	58.333333 % accepted	Seven accepted among twelve received.
One resolved	66.666667 % accepted	One existing unresolved record becomes accepted.
Two new unresolved	50.000000 % accepted	Two additional records arrive; accepted count stays seven.

CALCULATION LABORATORY

How do we reconcile the register without counting the same record twice?

Compare the cases on a fixed scale. Select a case to inspect its input change and calculated output.

Sensitivity case1. Current register2. One resolved3. Two new unresolved

Current register: 58.333333 % accepted. Seven accepted among twelve received.

Reproduce and challenge the result

Reconcile all twelve identifiers to the three exclusive states. Reject a snapshot in which accepted exceeds checked or checked exceeds received. Verify that re-deliveries update lineage instead of silently duplicating accepted coverage.

These are original synthetic calculations. The sensitivity cases vary the stated input while retaining the other assumptions; they are not field observations or treatment thresholds. The film’s calculation and sensitivity chapters use the same numerical cases as this table.

Find the exceptions in a full delivery register

The 60 by 200 matrix contains 12,000 synthetic receipt records. Codes 0, 1 and 2 mean unchecked, checked but unresolved, and accepted. The modular rules deliberately place reproducible gaps. They are data-generation rules, not an acceptance algorithm. Scan a row to find exceptions within a batch, then scan vertically to see repeated delivery patterns.

Reproduce the experiment. State = 0 if (i+3j) mod 29=0; else 1 if (i+j) mod 11=0; else 2. The downloadable table contains 12,000 evaluated cells, with the two input coordinates and output value retained on every row. These are parameter combinations, not that many independent field observations. The equation extends the worked example above; it is original teaching arithmetic, not a new standard or calibrated prediction.

Computed parameter atlas · 12,000 cells

Twelve thousand records, one awkward gap

A folder can look reassuringly full while the evidence register tells a different story.

Download every calculated cell (CSV) · Open full-size explorer · Equation source and tests (ZIP)

State codes are categories, not quantities: a height of two does not mean twice the quality. The display deliberately keeps those codes visible; do not average them into a quality score.

References and method notes

1. COTO (2016a). TMH13: Automated Road Condition Assessments. Part C: Roughness. Committee Draft Final, May 2016. Original source (Accessed: 5 September 2026).Sections C.2–C.5; Table C.5 is for calibration-section selection

Use the source edition and project requirements that apply to your work. Educational examples are not site-specific design advice.