Pavement Performance Management under Data Constraints

1

Section 1: Respondent Profile

Q1. What agency do you work for? *

Mark one option.

KeNHA KeRRA KURA KRB Other (specify)

Q2. At what level do you work? *

Mark one option.

Headquarters (national coordination) Regional office (regional work planning, implementation) Both Other (specify)

Q3. What position do you currently hold? *

Mark one option.

Director Deputy Director Assistant Director/Principal Engineer Senior Engineer Other (specify)

Q4. How many years of experience do you have in road asset management? *

Mark one option.

< 5 years 5–10 years 10–20 years > 20 years

Q5. What is your primary involvement? *

Select all that apply (at least one required).

Pavement condition assessment Maintenance planning/prioritisation Budget preparation and justification Road Management System (RMS) development and use/Data management Use of HDM-4 or other modelling tools Reporting to KRB/Ministry of Roads and Transport/Treasury Policy formulation Other (specify)

2

Section 2: Use of Condition Data in Decision-Making

3

Section 3: Continuity of Condition Records and Maintenance Histories

4

Section 4: Measurement Coverage and Data Gaps

5

Section 5: Forecasting and Deterioration Modelling

Q16. Does your agency use any formal method to forecast future pavement condition? *

Mark one option.

No formal forecasting Informal engineering judgement Yes – HDM-4 modelling for condition prediction and economic evaluation Yes – Structured statistical models (e.g., regression or trend-based models using historical condition data) Yes – Advanced data-driven models (e.g., machine learning or predictive models using large datasets) Other (please specify)

Q17. Do you agree or disagree with the following statement? "Current forecasting approaches reflect actual pavement deterioration patterns observed in the field." *

Mark one option.

Strongly agree Agree Neither agree nor disagree Disagree Strongly disagree

Q18. What are the biggest barriers to structured deterioration modelling? *

Select all that apply (at least one required).

Incomplete or inconsistent condition records Missing or unreliable maintenance histories Limited technical skills or modelling expertise Budget constraints Limited integration between data systems (condition, maintenance, planning) Difficult accessibility of historical records Institutional culture favouring established decision practices over analytical modelling Software/tool limitations Other (please specify)

Q19. If deterioration forecasts were demonstrably more reliable, how much impact would this have on the following planning outcomes? *

Mark one option in each row.

	No impact	Small impact	Moderate impact	Large impact	Very large impact
Multi-year budgeting
Treatment timing decisions
Network-level optimisation
Risk-based prioritisation
Justification of funding requests

6

Section 6: Data Reconstruction and Improvement

Q20. Where condition data are missing, how likely are you to consider the usage of analytical methods to estimate condition based on available network characteristics (e.g., road class, traffic level, pavement type, drainage condition, shoulder type, environmental and regional context)? *

Mark one option.

Very unlikely to consider Unlikely to consider Cannot say either way Likely to consider Very likely to consider

Q21. Under what conditions would you consider model-estimated or reconstructed condition data reliable for planning decisions? *

Select all that apply (at least one required).

Demonstrated agreement with measured field data Transparent modelling approach (clear methods and assumptions) Quantified uncertainty (e.g. confidence ranges) Independent technical review or validation Consistency with observed network behaviour (e.g., reconstructed values follow expected patterns across road classes, traffic levels, or known deterioration patterns) Approval by a recognized authority (e.g., KRB) Only when direct measurements are unavailable I would not consider such data reliable Other (please specify)

Q22. If model-estimated or reconstructed condition data were available with documented accuracy, how useful would they be for the following purposes? *

Mark one option in each row.

	Not useful at all	Slightly useful	Moderately useful	Very useful	Extremely useful
Network-level performance analysis
Life-cycle modelling and long-term planning
Treatment timing and optimisation
Budget allocation and justification

7

Section 7: Familiarity with Analytical Tools and Techniques

8

Section 8: Open-Ended Reflections

Q27. What practical improvements in data systems, institutional approaches, or technical capacity would most strengthen pavement performance management in Kenya? *

Provide your response in the text field.

Q28. Do you have any additional comments or recommendations regarding forecasting, modelling, or use of condition data in road asset management and planning? *

Provide your response in the text field.

Pavement Performance Management under Data Constraints

Section 1: Respondent Profile

Section 2: Use of Condition Data in Decision-Making

Section 3: Continuity of Condition Records and Maintenance Histories

Section 4: Measurement Coverage and Data Gaps

Section 5: Forecasting and Deterioration Modelling

Section 6: Data Reconstruction and Improvement

Section 7: Familiarity with Analytical Tools and Techniques

Section 8: Open-Ended Reflections

Thank You!