Kapil Kumar Meena

Kapil Kumar Meena

Postdoctoral ResearcherUniversity of California, Los Angeles

I am a postdoctoral researcher in the HUMAN Lab at the University of California, Los Angeles, working with Prof. Youngseo Kim. I study how people make travel decisions when the air is polluted, the day is hot or the information is new, and I build the methods and tools that turn this evidence into better routes, prices and policies.

I received my Ph.D. in Transportation Engineering from IIT Kharagpur in 2026, advised by Prof. Arkopal Kishore Goswami, and my M.Tech from IIT Roorkee, advised by Prof. Amit Agarwal. In 2024 I was a visiting research fellow at the Institute for Transport Studies, University of Leeds, and in 2021–22 a research intern at WRI India, working on commuter exposure and electric mobility.

Transportation How people choose modes and routes
Environment Pollution and heat on the road
AI & data science Interpretable models from sparse data

DRUM app The Hindu Published patent

1 published patent · 9 journal articles · 6 under review · 17 conference papers · reviewer for 8 journals

Email CV Google Scholar ORCID GitHub LinkedIn

Kapil Kumar Meena

News

2026

  • Paper

    Two papers accepted at TRB 2027, the 106th Annual Meeting in Washington, DC, in January 2027: seasonal air-quality preferences in mode choice and conflict screening from a single traffic image.

  • PatentJul

    Patent published. Our application on a personalised dynamic route planning system for sustainable urban mobility was published by the Indian Patent Office (No. 202631018379).

  • Career

    Joined the HUMAN Lab at UCLA as a postdoctoral researcher.

  • Career

    Completed my Ph.D. at IIT Kharagpur. Thesis: Assessing impacts of perceived air pollution exposure on urban travel behaviour.

  • TalkJan

    Presented at the 105th TRB Annual Meeting in Washington, DC: “Not all travellers think alike”, now accepted in Transportation Research Record.

  • Grant

    Institute travel grant (full funding) for the TRB Annual Meeting, IIT Kharagpur.

  • Paper

    Two conference papers on estimating air quality from sparse sensor networks, at IEEE COMSNETS and IEEE AIEI.

Research

Standard travel models describe choices mainly through time and cost. In Indian cities people also weigh the air they breathe, the heat they walk through, and whether they trust the information in front of them. I study these trade-offs with data I collect in the field, estimate them with methods I develop, and carry the results into tools and policy.

Field data

  • On-road PM2.5 sensing across transport micro-environments
  • Stated-preference experiments and two-wave commuter panels
  • Street and traffic images

Models

  • Discrete choice: hybrid latent class, ICLV, panel mixed logit
  • Convex estimation of nested and IPD logit models (PD‑MUSE)
  • Geostatistical deep learning; graphs with grounded LLMs

Decisions

  • DRUM routing app and a published patent
  • Pricing and allocating shade
  • Information design for EV charging
  • Evidence for clean-air and heat policy

Settings: air pollution, heat, electric mobility and road safety, in Delhi, Kolkata, Mumbai, Kharagpur and tier-2 and tier-3 Indian cities.

Figure 1. How the work fits together. Field data feed behavioural and machine-learning models, and the models feed tools, prices and policy.

Travel under air pollution and heat

I measure what commuters inhale across transport micro-environments (TPMDC 2022, J-EASTS 2025) and how they perceive it (JTH 2023). In a two-wave panel that followed the same 723 commuters in Kolkata from winter to summer, air quality carried about four times more weight in route choice in winter, and a pre-trip AQI alert was valued more than an equivalent cut in exposure (TBS, in revision). Hybrid latent class models separate commuters by how much exposure enters their decisions (TRR 2026). I am extending this to heat: how people value shade and where a city should provide it (Transport Policy, under review), and how heat changes when and how people travel (HEAT, at UCLA).

Information and decision tools

What travellers can do depends on what they know. I built DRUM, a routing engine that offers least-exposure and least-energy routes next to the shortest and fastest ones. In simulations for Central Delhi the least-exposure route cut exposure by half for 40% more travel time (TRR 2025). DRUM was featured in The Hindu and is the basis of a published Indian patent. Related work studies how information about public chargers shapes choice, reliability and trust in EV charging (TR-D, in revision), how air-quality awareness relates to travel (DAJ 2024), and how to encourage cycling to suburban rail and prioritise transit-oriented development (IJST 2025, STL 2025).

Estimation methods and behaviourally grounded AI

I develop methods that keep the structure of choice models while using what machine learning does well. PD‑MUSE estimates nested logit and related choice models by solving a convex problem and a small search, instead of one hard non-concave problem (manuscript in preparation). Other work joins econometric theory with deep learning for mode-choice prediction (TR-A, under review), estimates air quality from sparse sensor networks (COMSNETS 2026, AIEI 2026), and audits whether a single traffic image, read through a heterogeneous graph and a grounded LLM, can support conflict screening in mixed Indian traffic (TRB 2027, IATSS Research, under review).

Publications

A published patent, 9 journal articles, 6 manuscripts under review and 17 conference papers. Also on Google Scholar.

Patent

Media coverage

The Hindu, 8 June 2025: New app helps commuters pick greener routes on road
New app helps commuters pick ‘greener’ routes on road The Hindu 8 June 2025, by Ashmita Gupta

A feature on DRUM and the IIT Kharagpur team that built it, with results from the Delhi simulations.

Read onlinePrint page

Journal articles

Under review

In preparation

Conference papers

Software

Tools built from the research, most of them live.

Teaching and talks

Teaching interests

I can teach the core courses in transportation engineering, traffic engineering and transportation planning, and electives in travel demand and discrete choice analysis, transportation data analytics and machine learning, and transport, environment and health.

Teaching assistant

Invited talks

Service

Reviewer: Transport Policy, Transportation Research Part A, Transportation Research Record, Travel Behaviour and Society, Journal of Transport & Health, Research in Transportation Business & Management, Expert Systems with Applications, PLOS Climate.

Member: ASCE Transportation & Development Institute; Transportation Research Group of India; World Conference on Transport Research Society; International Association for Travel Behaviour Research.

Awards and grants