An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction

Narges Ahmadi; Yubo Jiao; Jônatas Augusto Manzolli; Jiangbo Yu; Luis Miranda-Moreno

cs.AIAug 20, 2026

An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction

Authors: Narges Ahmadi, Yubo Jiao, Jônatas Augusto Manzolli, Jiangbo Yu, Luis Miranda-Moreno

Organizations: Department of Civil Engineering, McGill University, Montreal, Quebec, Canada

Abstract

Travel behavior research increasingly combines digital data collection with predictive modeling, yet these stages are often developed and evaluated separately. This study proposes a three-agent workflow integrating conversational data collection, structured data processing, and behavioral prediction. A chatbot-administered, image-augmented stated-preference survey collected mode choices from student commuters across five predefined weather scenarios, yielding 454 respondent-scenario observations. Weather-related associations were analyzed using a multinomial logit model, while logistic regression and random forest provided machine-learning benchmarks. Nine locally deployed large language models (LLMs), ranging from 2 to 35 billion parameters, were evaluated across four zero-shot prompt-and-context conditions and extended through persona, few-shot, and vision-based configurations. Random forest achieved 69.6% five-class accuracy, while the best text-only zero-shot LLM reached 69.9% without task-specific fitting. Habitual travel information produced the most consistent gains, Expert framing generally outperformed Role-Play, and persona information was most useful when habitual travel information was unavailable. Few-shot prompting improved prediction for several models, with gains stabilizing after a small number of examples. Using the same weather images shown to respondents, the best vision-based configuration reached 71.5% five-class accuracy, indicating that visual context may provide additional predictive information for selected models. Overall, the study shows how conversational surveys, structured data processing, conventional behavioral modeling, machine learning, and multimodal LLM prediction can be coordinated within an auditable multi-agent workflow.

Figures & tables

Figure 1: Multi-agent framework for commuter travel-behavior research. Solid arrows denote forward data and artifact flows, whereas dashed arrows denote refinements implemented in a subsequent workflow iteration.

Agent	Main inputs	Authorized functions	Main outputs
Data Collection Agent	Survey specification and collection instructions	Chatbot construction, survey logic implementation, procedural question answering, response validation, clarification, deployment, and response aggregation	Deployable survey interface, distribution link or QR code, raw responses, interaction logs, and collection metadata
Data Processing Agent	Raw survey data, optional external data, codebook, and processing instructions	De-identification, validation, recoding, missing-data handling, text coding, data integration, feature construction, and quality assessment	Processed dataset, quality flags, derived variables, supplementary data, and processing records
Data Modeling Agent	Processed data, modeling instructions, candidate model library, and evaluation requirements	Model specification, code generation, statistical estimation, machine-learning training, LLM-based choice prediction, validation, and result organization	Fitted model, parameter estimates, predictions, uncertainty measures, diagnostics, evaluation metrics, and structured results

Table 1: Roles, interfaces, and supported functions of the three agents.

Panel of Figure 2 — Figure 2: Chatbot survey interface and sample weather-scenario question with embedded generated images.

Figure 3: Sample profile of survey respondents.

Figure 4: Stated mode share across weather scenarios (left) and change in mode share relative to the Sunny scenario, in percentage points (right).

Mode	Covariate	Coef.	Rob. t	p
Cycling	Rainy (vs. Sunny)	$-$ 0.856	$-$ 2.01	0.044*
Cycling	Foggy/cold (vs. Sunny)	$-$ 1.166	$-$ 2.52	0.012*
Cycling	Snowy (vs. Sunny)	$-$ 1.820	$-$ 2.63	0.009**
Cycling	Owns bicycle	1.219	2.06	0.039*
Cycling	Cycling experience (yrs.)	0.341	2.67	0.008**
Public Transit	Hot–humid (vs. Sunny)	0.518	2.17	0.030*

Table 2: Selected Multinomial Logit Coefficients (Reference Mode: Walking; Reference Weather: Sunny)

	EXP-RC		EXP-BC		RP-RC		RP-BC
Model	5-cl.	Binary	5-cl.	Binary	5-cl.	Binary	5-cl.	Binary
Gemma 4:e2B	44.0 $\pm$ 4.5	68.7 $\pm$ 4.1	31.6 $\pm$ 4.1	66.8 $\pm$ 4.3	41.2 $\pm$ 5.0	67.9 $\pm$ 4.2	32.3 $\pm$ 4.3	65.8 $\pm$ 4.1
Llama 3.2:3B	41.2 $\pm$ 4.3	62.5 $\pm$ 4.5	12.7 $\pm$ 3.6	56.7 $\pm$ 4.4	26.7 $\pm$ 3.7	60.7 $\pm$ 4.3	18.6 $\pm$ 3.6	60.3 $\pm$ 4.5
Gemma 3:4B	64.2 $\pm$ 4.4	79.1 $\pm$ 3.3	41.3 $\pm$ 4.0	64.1 $\pm$ 4.4	54.5 $\pm$ 4.2	75.9 $\pm$ 3.9	36.7 $\pm$ 4.2	55.6 $\pm$ 4.4
Gemma 4:e4B	42.8 $\pm$ 4.4	67.9 $\pm$ 4.4	40.4 $\pm$ 4.5	66.4 $\pm$ 4.2	40.1 $\pm$ 4.5	68.3 $\pm$ 4.0	35.7 $\pm$ 4.5	66.0 $\pm$ 4.2
Gemma 4:12B	69.9 $\pm$ 3.9	73.5 $\pm$ 4.2	64.6 $\pm$ 4.3	73.0 $\pm$ 4.2	69.8 $\pm$ 4.4	71.5 $\pm$ 4.1	63.9 $\pm$ 5.0	70.6 $\pm$ 4.1

Table 3: LLM Zero-Shot Five-Class and Binary Accuracy

Figure 5: Zero-shot accuracy across all four base conditions (Expert/Role-Play $\times$ Base/Richer Context) for nine models, shown separately for five-class and binary accuracy.

Panel of Figure 6 — Figure 6: Accuracy patterns by weather condition and mode across all evaluated models, with a bar-chart breakdown of weather- and mode-level accuracy under the EXP-RC experiment.

Figure 7: Combined LLM results. Top row: (left) zero-shot versus peak few-shot five-class accuracy per model, and (right) few-shot scaling curves across the six models. Bottom row: (left) experiment spread across all conditions, and (right) vision augmentation results comparing text-only baselines (EXP-RC and EXP-BC) against vision input (EXP-Img) and vision with three-shot prompting (EXP-ImgFS3).

Multimodal Large Language Models

Multimodal large language models combine language models with image, video, audio, and other inputs for grounded understanding and generation.

See all 1,092 papers

Fine-tuning LLMs for Tourist Trajectory Prediction using Field Experiment Data
Aug 21, 2026Tatsuya Amano, Hirozumi YamaguchiTrajectory Prediction LLM Fine-Tuning
The University of Osaka, Suita, Japan · RIKEN Center for Computational Science, Kobe, Japan
Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data
Sep 7, 2025Sepehr Golrokh Amin, Devin Rhoads, Fatemeh Fakhrmoosavi +2Synthetic Data Generation Synthetic Data Evaluation
School of Civil and Environmental Engineering, University of Connecticut, United States
LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals
Nov 15, 2024Joon Sung Park, Carolyn Q. Zou, Jonne Kamphorst +8Personality Modeling in Language Models Human Behavior Simulation
1Computer Science Department, Stanford University; Stanford, CA, 94305, USA. · Department of Communication Studies, Northwestern University; Evanston, IL 60208, USA. · 7CDSP and CEE, Sciences Po; Paris, 75007, France. +4

Aug 21, 2026J/K move · Enter open · S save

Fine-tuning LLMs for Tourist Trajectory Prediction using Field Experiment Data

Tatsuya Amano, Hirozumi Yamaguchi

The University of Osaka, Suita, Japan · RIKEN Center for Computational Science, Kobe, Japan

Trajectory Prediction LLM Fine-Tuning

Open paper AskPDF arXiv

Evaluating mobility interventions at tourist destinations requires predicting visitor behavior under varying conditions. Traditional methods struggle because tourist decisions depend heavily on context like weather and fatigue, yet models cannot generalize to unobserved scenarios. Large Language Models offer a solution by encoding commonsense knowledge about human behavior from pretraining, enabling reasoning about context-dependent decisions, while natural language representation flexibly integrates heterogeneous information. Fine-tuning on local trajectories adapts this general understanding to destination-specific patterns. We validate this approach using 566 trajectories from Wakayama Castle Park, Japan. Our fine-tuned Llama-3.1-8B achieves 49.1% next POI accuracy and maintains strong performance on undersampled scenarios like rainy days, demonstrating effective generalization. This establishes LLMs as high-fidelity behavior models for context-dependent tourist prediction, providing groundwork for counterfactual analysis of mobility interventions.

Abstract

Multimodal Large Language Models

Explore similar work

Fine-tuning LLMs for Tourist Trajectory Prediction using Field Experiment Data

Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data

LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals