Transportation & Traffic Engineering

Highly specialised requests are first checked for feasibility from this information. Please do not send confidential data through the public form; describe it first and agree secure transfer afterwards.

1. 01Enquiry and requirement assessment
2. 02Technical scoping and proposal
3. 03Execution
4. 04Quality review and delivery, with an agreed round of revisions

Read the full project workflow.

In depth: how the work is done

1. Traffic-flow and capacity analysis

   Fundamental relations among flow, density and speed describe how traffic behaves from free flow to congestion, with capacity as the maximum sustainable flow. Capacity analysis considers geometry, signal control and vehicle composition, using established methods and field calibration. We examine whether assumptions such as steady-state flow hold in the situation studied and how empirical data compare with theoretical relations.

2. Microscopic and macroscopic simulation

   Microscopic simulation models individual vehicles with car-following and lane-changing behaviour, and is suited to complex junctions, interactions and new control strategies. Macroscopic models treat traffic as a fluid, suitable for large networks and quick analysis. Each requires calibration to local conditions and validation against independent data, and the number of stochastic replications must be enough to estimate performance reliably. We describe model assumptions and the extent of calibration.

3. Statistical modelling of crashes and travel behaviour

   Crash data are counts that vary with exposure and are subject to under-reporting and regression-to-the-mean effects. Count models, such as negative binomial regression, and empirical Bayes methods account for overdispersion and for random fluctuation. Travel-behaviour studies use discrete-choice models to explain mode, route and departure-time choices. Model specification, data representativeness and interpretation of coefficients are examined so that conclusions do not exceed the evidence.

4. Before-and-after and network analysis

   Evaluating an intervention such as a signal retiming or a speed limit change requires comparing outcomes before and after, with controls for trends and external changes. Methods such as comparison groups or time-series interventions help separate the effect from background change. Network analysis, using graph measures and assignment models, addresses accessibility, resilience and flow redistribution after changes.

5. Mobility data analytics

   Probe, GPS, smart-card and mobile-device data provide detailed views of movement but have privacy implications, sampling biases and processing requirements. Map-matching, trip segmentation, expansion to the population and anonymisation are steps that influence results. We document the pipeline and its assumptions and advise on privacy and ethics considerations, noting that compliance with data-protection rules is the responsibility of the data holder.

6. Reporting traffic and mobility studies responsibly

   Transportation findings often influence public decisions, so reporting must separate what was measured from what was inferred. We recommend describing data coverage by time of day, location and vehicle type, stating how missing records were handled, and presenting uncertainty on every headline estimate. Where a model is used for forecasting, the report should list the scenarios assumed and avoid presenting a single forecast as certain. Sharing calibrated model files and processing scripts, where permitted, lets reviewers and later researchers reproduce the analysis and test alternative assumptions without rebuilding the study from scratch.

Common pitfalls this service helps avoid

• Ignoring regression to the mean. Sites selected for high crash counts often improve without intervention, overstating treatment effects.
• Uncalibrated simulation. Default behaviour parameters seldom represent local driving, and uncalibrated models can mislead.
• Biased samples. Probe or app data may over-represent some vehicles or travellers.
• Single stochastic run. Microscopic simulation outputs vary between runs and require multiple replications.

What good reporting looks like

A transportation report describes the study area, data sources with coverage and limitations, the models and their calibration and validation, results with confidence intervals or variability across replications, the evaluation design for interventions, the interpretation and limitations, and privacy considerations where relevant. It states which findings are descriptive and which causal, and it does not offer traffic design approvals or safety audits, which require qualified professionals and authorities.

An illustrative scenario

Hypothetical example written for explanation. It is not a client case, a result or a testimonial.

Imagine a research team studying a corridor where a signal coordination scheme was introduced. Travel-time data before and after show an apparent improvement, but a comparison corridor also improved due to a seasonal change in demand. A difference-in-differences analysis suggests a smaller but still positive effect for the treated corridor. The report presents the estimate with a confidence interval, discusses data coverage and notes that the effect on side-street delay was not measured.

Questions to settle before you begin

• What is the performance measure I need to evaluate?
• What data coverage and biases are present?
• How will I calibrate and validate any simulation?
• What comparison or control will isolate the effect of an intervention?
• How will I handle privacy and data permissions?

Key terms

Capacity

The maximum sustainable flow rate under given conditions.

Car-following model

A model describing how a driver responds to the vehicle ahead.

Negative binomial regression

A count model that allows for variance larger than the mean.

Empirical Bayes

A method combining site data and prior information to estimate expected crashes.

Discrete choice model

A model of the choice among a finite set of alternatives.

Working together on Civil, Structural & Infrastructure Engineering projects

Civil, structural and infrastructure research engagements start with a description of the structure or system, the loads and environment, the codes or standards that provide context and the questions the research must answer. We ask for drawings or descriptions, material data, monitoring data where they exist and any previous analyses. The work provides research analysis, literature-based assessment and documentation. It does not replace the design responsibility of licensed engineers, code compliance checks or site-specific professional judgement, and it states clearly where those are required.

Few projects need a single service in isolation. These are the services most often combined with this one.

• Engineering Data Science

  End-to-end analysis of engineering datasets, from cleaning and exploration to modelling and communication, with domain-aware handling of units, sensors and measurement artefacts.

• Regression & Predictive Modelling

  Development of regression and predictive models for engineering responses, from interpretable linear models to more flexible approaches, with validation designed to detect overfitting and extrapolation risk.

• Operations Research & Engineering Decision Models

  Mathematical-programming, simulation and decision-analysis models for scheduling, allocation, logistics and engineering decision problems under uncertainty.

• Industrial Engineering & Quality Management

  Analysis of production systems and quality processes using statistical quality control, process improvement frameworks and operations modelling.