Digital delivery for ageing infrastructure

From point cloud to a model an engineer can check

I lead digital delivery for ageing infrastructure: laser scans turned into models that can be checked, drawings turned into analysis models, design rules turned into automated checks, and condition assessment of ageing steel structures, including lattice steel towers and masts. The aim is always the same: a decision on a real asset that an engineer is prepared to sign.

Founding Dean, Smart City and Sustainable Development Academy (SCSDA), Chongqing Associate Professor (full-time), Southeast University Lead Guest Editor, Future Cities (Emerald) special issue “Renewing Ageing Infrastructure for Safe, Resilient and Liveable Future Cities” Guest Editor, Applied Sciences (MDPI) special issue on digital advancements in civil engineering and construction Selected as an Innovate UK Expert Assessor (October 2026) Member, ASCE (M.ASCE) · Member, IEEE · Member, CIGRE · Member, ASTM International · Affiliate, IStructE Former Associate Professor, OsloMet - Oslo Metropolitan University Dr.-Ing., Universitat der Bundeswehr Munchen Deputy Chair, Standards Digitalisation Committee, China Association for Engineering Construction Standardization (CECS)
What I Deliver How I Work Impact Record Research Agenda Publications
1,298Google Scholar citations
16h-index
24i10-index
90peer-reviewed papers

What I deliver

Six capabilities, each tested on real assets

Published methods and delivered work only. Client names, asset identifiers and monitoring data stay with the client.

Reality capture

Point clouds to models

Terrestrial laser scanning with target-based registration, through to scan-to-model. On a long-span bridge, a semi-automated pipeline turned the point cloud into editable CAD drawings, with up to 91% overlap (IoU) on critical structural outlines and 83% of key elements identified automatically, and the result was used to update the finite element model. In a study I co-authored, virtual trial assembly of scanned steel components came within 1/1800 of span of the physical assembly. My current work links this to UAV LiDAR and imagery for lattice towers.

Drawings to analysis

Models that can be checked

CAD drawings parsed by graph methods into finite element models, run end to end with reconstructed wind fields, and gates that record where each input came from: source time, spatial extent and solver balance (Applied Sciences, 2026).

Automated checking

Rules that software can check

Code clauses translated into executable rules for automated BIM compliance checking, with a person in the loop: 95.8% accurate rule translation, 98.3% executable, and checking effort on a real project cut from 1,620 to 168 hours (Buildings, 2026).

Condition assessment

Ageing steel structures

Reliability of ageing lattice steel towers and masts under extreme wind and corrosion: image-based corrosion grading, measured section loss, slipped connections and lost preload. For a 40-year-old coastal tower, corrosion rate and zinc coating thickness together govern 65% of reliability variance (Buildings, 2026).

Asset information

Digital twins that are kept up to date

Twins updated from measurement rather than drawn once, with information requirements and data structures aligned with ISO 19650, so that the next team, with different tools, can re-derive the same conclusion from the same data.

Delivery

Teams, programmes and repeat clients

Staged programmes in which each stage leaves the client something usable before the next begins. I have led teams of twelve to fifteen across modelling, sensing, AI and physical testing, and one client has commissioned three successive programmes from me, each wider in scope than the last.

Working with

Collaboration

Editorial

Future Cities special issue

Co-edited with Dr Long Chen (City University of Hong Kong) and Dr Pingbo Tang (Carnegie Mellon University), with the support of Professor Stergios Mitoulis (UCL), founding Editor-in-Chief of Future Cities. Submissions open on 1 November 2026 and close on 28 February 2027; papers submitted by 31 December 2026 pay no article processing charge. Call for papers.

Research

Joint papers and proposals

In discussion with colleagues at the universities of Cambridge, Manchester and Bristol, Loughborough University, TU Delft and NTNU on joint papers and proposals in structural assessment, reality capture and infrastructure resilience.

Research at a glance

How does measured condition guide the safe life extension of structures in service?

Six lines of work answer that question from different sides: assessment and life extension, resilience under extreme loading, monitoring by scanning, vibrometry and MEMS, digital twins with model updating, low-carbon and recycled concrete, and automation with machine-readable codes.

Sensing on cable-supported bridges
Sensing on cable-supported bridgesMEMS strain, temperature and vibration nodes installed on two cable-supported bridges during construction.
Measured geometry of a long-span arch
Measured geometry of a long-span archLaser scanning of a mountain arch bridge through construction, turned into an auditable structural model.
Structures under extreme loading
Structures under extreme loadingReinforced concrete members under impact: finite element failure modes checked against test.
Automation testbed
Automation testbedA 1:10 intelligent building-machine platform at Southeast University, signed off against 17 criteria.
Virtual pre-assembly
Virtual pre-assemblyLaser-scanned steel components fitted together in the model before they reach site.
Construction waste as material
Construction waste as materialExcavation and demolition waste processed into recycled aggregate for road rehabilitation.
Monitoring hardware
Monitoring hardwareA wireless sensing node developed for long-term structural monitoring.
Sub-millimetre scanning
Sub-millimetre scanningTerrestrial laser scanning with registration targets, the first step of every twin I build.

How I work

Four layers, each harder to replicate than the one above it. The fourth is the one that decides whether the first three ever reach an asset.

Layer 1 and 2

Knowledge, software and standards

Hazard-structure coupling, failure probability and residual capacity, topology optimisation, data governance. Above that sit the things that travel: digital twin platform architecture, a data-asset meta-model and access standard for ageing assets, inspection procedures and draft evaluation standards that do not depend on the original data.

Layer 3

Hardware that has been built and tested

Replaceable steel connections, modular emergency tower design, climbing inspection and assembly robotics, and before-and-after strengthening trials on decommissioned structures. Most comparable work stops at proof of concept and papers.

Layer 4

Organisation and consortium design

Research groups join with their own teams and plug into a shared framework of directions, targets, funding and facilities. The projects they win sign through the platform, while a small in-house team runs finance, delivery, market development and commercialisation. Centres established with partner institutions in Europe run the same model, with the platform paying the partner side. What holds an arrangement like this together over years is not goodwill but paperwork: every centre rests on a dated, signed agreement rather than an informal contact, which is what lets a partner commit people to it.

Nobody was on my payroll. The arrangement did the work. Three digital capability centres have been built from zero this way, and in each case none of the disciplines involved reported to me. That is not a limitation of the method; it is the method.

Evidence and assurance

How claims are handled when the work is technical, the client is accountable, and part of the delivery is AI-assisted.

What gets written down

Failures are recorded alongside passes

Where a benchmark was not met, it goes into the stage report next to the tests that passed, with the calibration work that follows. Where a target is contractual rather than achieved, we say so: an acceptance target is not a realised outcome until it has been checked against real cases.

AI-assisted delivery

Every claim traces to source material

On AI-assisted deliverables, each statement traces back to material the client provided. The model does not get to find its own facts. Assurance is the product, not the packaging.

Selected Research Evidence

Where the research has been examined, taught, standardised and put to work.

Publications

Digital twin predictive maintenance framework

High-impact work on digital twin-enabled predictive maintenance for building operations, contributing to asset management and infrastructure data research.

co-author411 citationsEnergy and Buildings

Standards

Standardise the data, not the algorithm

Algorithms are replaced continuously and are dated by the time they reach the field. Complete, consistent, standardised data is the shared input every future generation of algorithm will need, which is where the durable asset sits. I work on this from inside the standards process, as deputy chair of the standards digitalisation committee of the China Association for Engineering Construction Standardization, and in my own research on machine-readable codes: a 2026 paper turns written clauses into executable checks with a person in the loop.

CECS deputy chairmachine-readable codes

Teaching

English-medium and project-based teaching

Former OsloMet Associate Professor experience, postgraduate teaching in BIM and digital engineering, Loughborough guest teaching, and supervision of doctoral and master's research.

doctoral and master's supervisionEnglish-medium teaching

Knowledge transfer

Research reaching engineering practice

Published methods written into industry inspection procedures, commissioned resilience research for large infrastructure operators, and a research platform built with district and municipal support in Chongqing, reported independently by national and regional media.

Impact recordstatements available on request
What appears here is the published and publicly verifiable part of the work. Client identities, commercial terms, unpublished technical parameters and work still under review are not discussed in public, and are not discussed privately either without the agreement of the parties concerned.