Syed Nafiz Reja Hasib

Civil and structural engineer

Contact
005

Climate-driven moisture damage in nano-modified warm mix asphalt

A small, transparent scenario model that propagates a laboratory moisture-durability indicator, the tensile strength ratio, through time under monthly temperature and rain, and writes outputs that a BIM or asset record can store.

Year
2026
Status
Complete
Methods
Durability modelling, Validation
Tools
Python, NumPy, pandas, Matplotlib
Diagram of the model: monthly climate series, exposure index, tensile strength ratio decay and BIM-ready output tables.
Fig. 0Fig. 1, model architecture: monthly climate → exposure index → TSR decay → BIM-ready records.

Key numbers

Climate ensemble
2,000realisations
Months to 60/70/80 % TSR reported as median and 5–95 % range
Primary run
60months
Seeded; reproduced exactly on a newer Python stack

Question

Laboratory moisture tests give one number at one moment. How could that number be carried forward in time, under a real climate, into something an asset manager can store and act on?

Approach

Each month gets a dimensionless exposure index, E = W(P) · Q10^((T − T_ref)/10), where W(P) is a saturating wetness function equal to 1 at the reference precipitation. The tensile strength ratio (TSR) relaxes toward a residual value, dTSR/dt = −α · E · (TSR − TSR_res), solved exactly month by month. A nano-modified mixture is represented by a higher initial TSR and a lower decay coefficient.

The model writes tables and figures that an asset database or BIM record could store, and includes a least-squares routine to re-calibrate the decay coefficient from measured core TSR values.

Validation

This is an empirical scenario model, not a validated predictor. What is checked is the engine: a built-in self-test, exact month-by-month solution, and full reproducibility (a re-run on a newer Python, NumPy, pandas and Matplotlib reproduced every reported value).

Results

Months to the 60, 70 and 80 % TSR thresholds are reported as a median and 5–95 % range over 2,000 climate realisations, with one-at-a-time sensitivity of every assumption.

What this does not show

  • All material parameters are scenario inputs, and the climate series is synthetic around published monthly normals.
  • Outputs are not service-life predictions.
  • In the terms of Kritzinger et al. (2018) this is a digital model: no data flow to or from a physical pavement is automated.

Evidence

Code
github.com/syednafizreja/pavement-digital-twin
Tests
Built-in self-test (--selftest)
Licence
MIT
Tensile strength ratio over time for standard and nano-modified mixtures under the climate ensemble.
Fig. 3, TSR over time for both mixtures under the climate ensemble (2,000 realisations).
004

PaveCarbon

An open finite-element and life-cycle framework that carries a laboratory durability gain of nano-silica warm mix asphalt through a pavement structure to service life and annualised A1–A3 carbon, so the carbon claim becomes auditable.