Pith. sign in

REVIEW 4 major objections 5 minor 33 references

Beyond 9-to-5: A Generative Model for Augmenting Mobility Data of Underrepresented Shift Workers

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims a transformer-based generative model can turn fragmented GPS traces of shift workers into complete activity chains whose population-level distributions match observed Los Angeles GPS data to within 0.02 Jensen–Shannon…

desk verdict A useful shift-worker mobility application whose headline JSD numbers are partly trained-for and possibly leakage-inflated; the paper is worth reviewing but needs agent-disjoint splits and independent label validation. read the letter →

arxiv 2507.19510 v1 pith:NMDM55TA submitted 2025-07-17 cs.LG cs.AI

Haoxuan Ma , Xishun Liao , Yifan Liu , Chris Stanford , Jiaqi Ma This is my paper · ORCID
classification cs.LGcs.AI
keywords shiftworkersactivitychaingenerationGPStrajectoryimputationtransformerperiod-awaretemporalembeddingtransition-awarelossJensen-Shannondivergencemobilitydataaugmentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Shift workers make up 15–20% of the workforce, but conventional household travel surveys systematically miss their overnight and midnight-crossing activity, so planners plan around a 9-to-5 city. This paper proposes a transformer-based generative model that takes fragmented GPS traces of shift workers and fills in complete, behaviorally plausible activity chains. The central claim is that the generated chains match real GPS-derived shift-worker patterns in Los Angeles County at the population level, with an average Jensen–Shannon divergence of 0.0176 across temporal and activity-type distributions. If true, the model is a data-augmentation tool that lets planners correct survey bias without new data collection.

What carries the argument

The core mechanism is a transformer encoder–decoder that ingests an embedded activity chain plus an observation mask and autoregressively generates the next day's activities one 15-minute time slot at a time. Three design pieces carry the argument: period-aware temporal embeddings that split the day into evening-start, overnight, morning, and other periods; a transition-aware loss that scores precision and recall of activity boundaries within a tolerance window of two time slots; and a distribution-matching loss minimizing Jensen–Shannon divergence between generated and real activity-type frequencies. A progressive masking schedule during training teaches the model to infer activities through the gaps that characterize GPS coverage loss.

What would settle it

Take a held-out sample of shift-worker GPS traces, have the participants record their actual activities in a diary for the same days, and compare the model's generated activity chains to those diary records with the same JSD metrics; if the divergence is much larger than the reported ~0.02, the low JSD reflects alignment to the labeling pipeline rather than behavioral validity.

Watch

Extended reading notes

Core claim

The paper claims that a transformer encoder–decoder, trained on pairs of consecutive days from 208,350 shift-worker activity sequences, can reconstruct a complete next-day activity chain from a partially observed day, including activity types, start/end times, and durations. Evaluated against held-out GPS-derived activity chains from Los Angeles County, the generated chains have an average JSD of 0.0176 across start-time, end-time, duration, and activity-type distributions, compared with 0.0621 for an LSTM with attention baseline. The model reproduces the distinctive bimodal work pattern of shift workers—peaks around midnight and late evening—that normal-worker and NHTS survey distributions miss. The paper presents this as evidence that incomplete GPS traces can be converted into complete activity diaries that are representative of shift-worker mobility for transportation planning.

Load-bearing premise

The activity labels that serve as ground truth—built by extracting stay points from GPS and labeling the places with an LLM-informed POI classifier—are assumed to accurately reflect what shift workers actually do.

Editorial extensions

If this is right

  • Transportation planners can use the model to turn partial GPS traces into complete shift-worker activity chains, filling the overnight hours that surveys miss.
  • Augmenting household travel survey data with generated shift-worker patterns would rebalance the systematic underrepresentation of the 15–20% of workers on non-standard schedules.
  • Because the model reproduces bimodal overnight work peaks, downstream analyses of late-night transit demand and service scheduling can be built on distributions that match GPS observations.
  • The framework transfers to other cities without architectural changes because it relies only on GPS and point-of-interest data for training.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension is to apply the same masked-sequence generation to other under-surveyed populations, such as night-time transit users or caregivers, where the same survey bias operates.
  • The reported JSD measures agreement with the GPS-derived labels; an independent validation against time-use diaries or manual inspection would determine whether the generated chains are behaviorally valid, not merely label-consistent.
  • A direct policy test would be to compare late-night transit service plans produced from survey-only data versus survey data augmented with generated shift-worker chains, to quantify how much the bias correction changes investment priorities.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a transformer encoder-decoder model that reconstructs complete two-day activity chains for shift workers from partially observed GPS-derived activity sequences. The motivating observation is that conventional household travel surveys undersample non-standard-hour workers, while GPS traces are fragmented. The method uses period-aware temporal embeddings (evening, overnight, morning), a progressive masking schedule to simulate GPS gaps, and a composite loss consisting of cross-entropy, transition-aware F1, Jensen-Shannon divergence between generated and real activity-type distributions, and a soft-label transition term. Evaluation is performed on 208,350 two-day activity sequences extracted from six months of Los Angeles County GPS data, with an 80/10/10 sequence split. The headline result is an average JSD of 0.0176 against real GPS distributions, compared to 0.0621 for an LSTM-with-attention baseline, and the paper claims this demonstrates behaviorally valid generation suitable for data augmentation in transportation planning.

Significance. If the central claim were fully established, the paper would make a useful contribution to activity-based travel demand modeling: it targets an important and genuinely underserved population, proposes a sensible way to handle fragmented GPS observations via masking, and introduces period-aware embeddings and a transition-sensitive loss that are well motivated by the shift-worker setting. The comparative analysis of HTS versus GPS start-time distributions (Table II) is a valuable empirical illustration of survey bias, and the paper explicitly frames the model as a data-augmentation tool rather than a next-trip predictor, which is the right framing. However, the headline distributional-alignment claim is not yet supported by the evidence as presented: the evaluation metric overlaps with a training objective, the data split is by sequence rather than by agent, the activity labels used as ground truth are unvalidated, and there are no error bars or significance tests. These are load-bearing gaps for the paper's central claim that the model generates behaviorally valid activity chains for new shift workers.

major comments (4)
  1. [§IV.C.3 and §V.D, Table III] The evaluation metric is not independent of the training objective. The distribution-matching loss Ldist = JS(P_Ŷ || P_Y) in §IV.C.3 directly minimizes the Jensen-Shannon divergence between generated and real activity-type distributions, and Table III then reports JSD against the same real distribution as the headline metric. The Activity Type row (JSD 0.0049) is therefore partly a property of the training loss, not an external measure of generalization or behavioral validity. This does not invalidate the method, but it means the claim 'Average JSD < 0.02 for all evaluation metrics' cannot be read as a fully independent benchmark. I recommend reporting at least one evaluation statistic that is not present in any loss component (for example, transition timing JSD, sequence-level entropy, or a separate held-out distributional moment), and ideally ablating the model with β=0 to show the marginal contribution of Ldist.
  2. [§V.B] The data split is described only as an 80/10/10 split of 'sequences', with no statement that sequences from the same agent are kept in the same split. If the 208,350 two-day sequences come from a smaller number of agents, a random sequence split places the same person's day-1/day-2 pairs in both training and test sets. Because the model is conditioned on day 1 and asked to generate day 2, a worker with a repeating shift schedule could be near-memorized, producing artificially low JSD that does not reflect generalization to new shift workers. This is load-bearing for the claimed data-augmentation utility. Please report the number of agents, describe how the split was performed, and either provide an agent-disjoint split or demonstrate that the results are unchanged when sequences from the same agent are grouped into one split.
  3. [§V.A.b and Limitations] The ground-truth activity labels are produced by stay point extraction and LLM-informed POI classification referenced to [33], and the paper provides no validation of these labels against travel diaries, surveys, manual inspection, or any other reference. The reported JSD values measure alignment with this unvalidated proxy, not with real shift-worker behavior. The Limitations section acknowledges that 'activity detection relies on GPS-POI matching with rule-based assumptions', which is honest, but the abstract and conclusion state the model generates 'behaviorally valid activity patterns' without qualification. Please add a label-validation subsection or a sensitivity analysis (e.g., agreement rates on a manually annotated subset, or an ablation with alternative POI-labeling thresholds), and soften the behavioral-validity claim until such validation exists.
  4. [§V.D, Table III] No error bars, confidence intervals, or significance tests are reported, and the comparison is against a single baseline (LSTM with attention) with no unconditional marginal-frequency predictor. Without repeated runs or a statistical test, the difference between 0.0176 and 0.0621 cannot be assessed for significance, and a trivial baseline that simply predicts the population marginal activity-type distribution would clarify how much of the low JSD comes from capturing marginals versus temporal structure. Please report means and standard deviations over at least three random seeds, add a marginal-baseline row to Table III, and state whether the train/test metrics differ materially under an agent-disjoint split.
minor comments (5)
  1. [§V.C] The paragraph beginning 'We evaluate our model by comparing the distributions of various mobility characteristics...' is duplicated verbatim in the evaluation methodology section; please remove the duplicate.
  2. [§I, Contributions] The first contribution bullet contains 'rebalancing systemic undersampling in plannings'; 'plannings' should be 'planning'.
  3. [§IV.B.1] The period function p(t) uses 'other' as a catch-all period, but it is not clear how the transition between 'other' and the named periods is embedded; a sentence explaining how the period embedding is chosen for t outside the listed intervals would improve reproducibility.
  4. [§V.A.b] The description of shift-worker identification criteria (evening work 18:00-22:00, midnight-crossing work, sustained work 22:00-06:00) would benefit from the exact thresholds used (e.g., minimum duration to count as 'sustained', or whether criteria are applied per day or per sequence), since this directly affects the composition of the dataset.
  5. [§V.D.1] The interpretation of the duration JSD (0.0411) attributes the discrepancy to GPS fragmentation artifacts, but this claim is not directly evidenced; consider reporting a comparison of generated duration distributions against the empirical distribution of activities with confirmed, uninterrupted GPS coverage.

Circularity Check

1 steps flagged · score 6.0 of 10

The reported activity-type JSD is the same divergence minimized by the Ldist training term, so part of the headline alignment is forced by construction.

  1. fitted input called prediction [Section IV.C.3 (Distribution Matching Loss); evaluated in Section V.D.1, Table III; abstract]
    "To ensure the overall distribution of activities is realistic, we include a distribution matching component based on JSD ... Ldist = JS(P_Yhat || P_Y). ... Table III reports Activity Type JSD = 0.0049, and the abstract claims 'Average JSD < 0.02 for all evaluation metrics'."

    The activity-type distribution JSD in Table III is exactly the quantity Ldist minimizes during training. The model's generated marginal activity-type distribution is explicitly pushed toward the real distribution by the loss, so reporting this JSD as evidence of distributional validity is measuring the fitted objective, not an independent prediction. The headline average JSD (0.0176) includes this directly optimized component, so the central 'remarkable distributional alignment' claim is partially forced by construction. Start/end time, duration, and work-specific JSDs are not explicit terms in Ldist, so the circularity is partial.

full rationale

The paper's main derivation chain is otherwise self-contained: the transformer, masking, and period-aware embeddings are architectural choices; the start-time, end-time, duration, and work-specific JSDs are not directly minimized as JSD terms and retain independent content. I do not count the LLM-POI activity labels from the authors' prior work [33] as circular, because they are produced by an external labeling pipeline and the paper explicitly concedes in Limitations that GPS-POI matching introduces potential biases. The absence of an agent-disjoint split is a generalization/leakage risk, not a definitional circularity. The one construction-level problem is the use of Ldist as a training objective followed by evaluation on the same JS divergence for activity type, which inflates the headline average JSD.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The model introduces no physical entities. It relies on several domain assumptions: GPS-POI labels are valid ground truth, shift workers are correctly identified by temporal work criteria, and distributional JSD is an adequate validity measure. The main free parameters are loss weights, transition tolerance, period boundaries, shift worker criteria, and the masking schedule, none of which are optimized or sensitivity-tested.

free parameters (7)
  • Loss weights alpha, beta, gamma (Lcombined = LCE + alpha*Ltrans + beta*Ldist + gamma*Lsoft) = not reported
    Configurable weights for the combined loss; their values are never stated, so the balance between distribution matching and transition accuracy is unspecified and not sensitivity-tested.
  • Transition tolerance tau = 2 time slots (30 minutes)
    Tolerance window for transition-aware loss; chosen by hand, no sensitivity analysis.
  • Period boundaries for period-aware embedding = evening 18-22, overnight 22-06, morning 06-10
    Piecewise period definition is a hand-set modeling choice that directly shapes the temporal representation of shift workers.
  • Shift worker identification thresholds = work in 18-22, midnight crossing, work in 22-06
    These three criteria define the study population; no validation of their sensitivity or external comparison to known shift worker rosters.
  • Progressive masking schedule = not reported
    Training increases masking ratio over time to simulate GPS gaps, but the schedule, starting ratio, and endpoint are unspecified.
  • Model hyperparameters = 4 layers, 8 heads, dim 128, lr 1e-4, weight decay 1e-5, dropout 0.1, 50 epochs, batch 256
    Chosen by hand; standard values, but the paper reports no ablation or seed variation.
  • Time discretization = 96 slots of 15 minutes
    Coarse temporal resolution is a modeling choice; fine-grained overnight patterns may be blurred.
assumptions (6)
  • standard math Cross-entropy, JSD, and transformer backpropagation are valid statistical and optimization tools.
    Used throughout Sec IV without proof; standard results.
  • domain assumption GPS stay point extraction and LLM-informed POI classification produce activity labels that approximate true activities.
    All training and evaluation relies on these labels; no independent validation is provided (Sec V.A.b, Limitations).
  • domain assumption Shift workers can be identified by evening/overnight work timing criteria.
    The study population is defined by three temporal rules in Sec V.A.b; no ground truth roster or survey is used.
  • domain assumption A pair of consecutive daily activity chains contains enough information to predict the next day's chain.
    The problem formulation and model use day 1 to generate day 2; this assumes temporal continuity of activity behavior across days.
  • domain assumption Distributional JSD over temporal and activity-type distributions is an adequate measure of behavioral validity.
    The evaluation claims behaviorally valid patterns based on JSD; no downstream planning task or individual-level accuracy is measured.
  • domain assumption The NHTS activity taxonomy [32] covers shift worker activities without systematic omission.
    Activities are mapped into 15 categories; unusual shift worker activities may be forced into 'Other'.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Beyond 9-to-5: A Generative Model for Augmenting Mobility Data of Underrepresented Shift Workers." pith.science (2026). https://pith.science/paper/NMDM55TA

@misc{pith2026250719510,
  author       = {Pith},
  title        = {Pith review of: Beyond 9-to-5: A Generative Model for Augmenting Mobility Data of Underrepresented Shift Workers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NMDM55TA}},
  note         = {Machine review of arXiv:2507.19510}
}
read the original abstract

This paper addresses a critical gap in urban mobility modeling by focusing on shift workers, a population segment comprising 15-20% of the workforce in industrialized societies yet systematically underrepresented in traditional transportation surveys and planning. This underrepresentation is revealed in this study by a comparative analysis of GPS and survey data, highlighting stark differences between the bimodal temporal patterns of shift workers and the conventional 9-to-5 schedules recorded in surveys. To address this bias, we introduce a novel transformer-based approach that leverages fragmented GPS trajectory data to generate complete, behaviorally valid activity patterns for individuals working non-standard hours. Our method employs periodaware temporal embeddings and a transition-focused loss function specifically designed to capture the unique activity rhythms of shift workers and mitigate the inherent biases in conventional transportation datasets. Evaluation shows that the generated data achieves remarkable distributional alignment with GPS data from Los Angeles County (Average JSD < 0.02 for all evaluation metrics). By transforming incomplete GPS traces into complete, representative activity patterns, our approach provides transportation planners with a powerful data augmentation tool to fill critical gaps in understanding the 24/7 mobility needs of urban populations, enabling precise and inclusive transportation planning.

Figures

Figures reproduced from arXiv: 2507.19510 by the authors.

Figure 2
Figure 2. Network Architecture of the Model [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 1
Figure 1. System Workflow of the Proposed Model to evening and overnight periods. We processed the GPS points and constructed corresponding masking tensors to address the inherent gaps in GPS data collection, ensuring the model learns from reliable observations while appropriately handling missing data. Our training pipeline incorporates specialized data aug￾mentation techniques that emphasize realistic overnight ac￾tivity tr… view at source ↗
Figure 3
Figure 3. Comparison of distributional characteristics between generated and [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Comparison of work activity patterns between different worker [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 32 canonical work pages

  1. [33]

    Semantic Trajectory Data Mining with LLM-Informed POI Classification

    Y . Liu, C. Kuai, H. Ma, X. Liao, B. Y . He, and J. Ma, “Semantic trajectory data mining with llm-informed poi classification,” arXiv preprint arXiv:2405.11715, 2024

    work page Pith review arXiv 2024

  2. [1]

    Understanding individual human mobility patterns,

    M. C. Gonzalez, C. A. Hidalgo, and A.-L. Barabasi, “Understanding individual human mobility patterns,” Nature, vol. 453, 2008

    work page 2008

  3. [2]

    Deep activity model: A generative approach for human mobility pattern synthesis,

    X. Liao, B. Y . He, Q. Jiang, C. Kuai, and J. Ma, “Deep activity model: A generative approach for human mobility pattern synthesis,” arXiv preprint arXiv:2405.17468, 2024

    arXiv 2024

  4. [3]

    Numosim: A synthetic mobility dataset with anomaly detection benchmarks,

    C. Stanford, S. Adari, X. Liao, Y . He, Q. Jiang, C. Kuai, J. Ma, E. Tung, Y . Qian, L. Zhao et al. , “Numosim: A synthetic mobility dataset with anomaly detection benchmarks,” in Proceedings of the 1st ACM SIGSPATIAL International Workshop on Geospatial Anomaly Detection, 2024, pp. 68–78

    work page 2024

  5. [4]

    Spa- tiotemporal patterns of urban human mobility,

    S. Hasan, C. M. Schneider, S. V . Ukkusuri, and M. C. González, “Spa- tiotemporal patterns of urban human mobility,” Journal of Statistical Physics, vol. 151, no. 1, pp. 304–318, 2013

    work page 2013

  6. [5]

    Mobility AI agents and networks,

    H. Ma, Y . Liu, Q. Jiang, B. Y . He, X. Liao, and J. Ma, “Mobility AI agents and networks,” IEEE Transactions on Intelligent Vehicles , vol. 9, no. 7, pp. 5124–5129, 2024

    work page 2024

  7. [6]

    A two-sided model for ev market dynamics and policy implications,

    H. Ma, B. Y . He, T. Kaljevic, and J. Ma, “A two-sided model for ev market dynamics and policy implications,” in 2024 IEEE 27th International Conference on Intelligent Transportation Systems , 2024

    work page 2024

  8. [7]

    Sustainable intelligent transportation systems via digital twins: A contextualized survey,

    V . M. G. Martinez, D. R. Campelo, and M. R. N. Ribeiro, “Sustainable intelligent transportation systems via digital twins: A contextualized survey,” IEEE Open Journal of Intelligent Transportation Systems , vol. 6, pp. 363–392, 2025

    work page 2025

Show all 33 references
  1. [8]

    Sixth european working conditions survey – overview report,

    A. Parent-Thirion, I. Biletta, J. Cabrita, O. Vargas Llave, G. Ver- meylen, A. Wilczy ´nska, and M. Wilkens, “Sixth european working conditions survey – overview report,” European Foundation for the Improvement of Living and Working Conditions, Tech. Rep., 2016

    2016

  2. [9]

    Long workweeks and strange hours,

    D. S. Hamermesh and E. Stancanelli, “Long workweeks and strange hours,” National Bureau of Economic Research, Tech. Rep., 2014

    2014

  3. [10]

    Supporting late-shift workers: Their transportation needs and the economy,

    American Public Transportation Association, “Supporting late-shift workers: Their transportation needs and the economy,” American Public Transportation Association, Tech. Rep., 2019

    2019

  4. [11]

    The impact of night and evening shift work on social exclusion, family travel, and mobilities of care,

    M. Palm, “The impact of night and evening shift work on social exclusion, family travel, and mobilities of care,” Center for Open Science, SocArXiv, 2024

    2024

  5. [12]

    Shifted out: The well-being and justice implications of evening and night commuting,

    M. Palm, J. Allen, and S. Farber, “Shifted out: The well-being and justice implications of evening and night commuting,” Transportation Research Part D: Transport and Environment , 2023

    2023

  6. [13]

    2017/2018 regional travel survey in-depth analysis: Late-night travel,

    N. C. R. T. P. Board, “2017/2018 regional travel survey in-depth analysis: Late-night travel,” Tech. Rep., 2019

    2017

  7. [14]

    Tourangeau, Defining hard-to-survey populations

    R. Tourangeau, Defining hard-to-survey populations . Cambridge University Press, 2014

    2014

  8. [15]

    Investigating night shift workers’ commuting patterns using passive mobility data,

    S. Lim, H. Ahn, S. Shin, D. Lee, and Y . H. Kim, “Investigating night shift workers’ commuting patterns using passive mobility data,” Transportation Research Part A: Policy and Practice , vol. 181, p. 104002, 2024

    2024

  9. [16]

    Agent-based models in urban transportation: review, challenges, and opportunities,

    F. F. Bastarianto, T. O. Hancock, C. F. Choudhury, and E. Manley, “Agent-based models in urban transportation: review, challenges, and opportunities,” European Transport Research Review, 2023

    2023

  10. [17]

    C. R. Bhat and F. S. Koppelman, Activity-based modeling of travel demand. Springer, 1999, pp. 35–61

    1999

  11. [18]

    Activity-based disaggregate travel demand model system with activity schedules,

    J. L. Bowman and M. E. Ben-Akiva, “Activity-based disaggregate travel demand model system with activity schedules,” Transportation Research Part A: Policy and Practice , vol. 35, no. 1, pp. 1–28, 2001

    2001

  12. [19]

    Analysis of human mobility patterns from gps trajectories and contextual information,

    K. Siła-Nowicka, J. Vandrol, T. Oshan, J. Long, U. Demsar, and A. S. Fotheringham, “Analysis of human mobility patterns from gps trajectories and contextual information,” International Journal of Geographical Information Science, vol. 30, no. 5, pp. 881–906, 2016

    2016

  13. [20]

    New approaches to human mobility: Using mobile phones for demographic research,

    J. R. B. Palmer, T. J. Espenshade, F. Bartumeus, C. Y . Chung, G. A. Thakur, and J. R. Goldstein, “New approaches to human mobility: Using mobile phones for demographic research,” Demography, 2013

    2013

  14. [21]

    A reference architecture for data-driven intelligent public trans- portation systems,

    F. Rocco Di Torrepadula, S. Di Martino, N. Mazzocca, and P. Sannino, “A reference architecture for data-driven intelligent public trans- portation systems,” IEEE Open Journal of Intelligent Transportation Systems, vol. 5, pp. 469–482, 2024

    2024

  15. [22]

    Multi-hop upstream anticipatory traffic signal control with deep reinforcement learning,

    X. Li, X. Wang, I. Smirnov, S. Sanner, and B. Abdulhai, “Multi-hop upstream anticipatory traffic signal control with deep reinforcement learning,” IEEE Open Journal of Intelligent Transportation Systems , vol. 6, pp. 554–567, 2025

    2025

  16. [23]

    Integrating multimodality and partial observability solutions into decentralized multiagent reinforcement learning adaptive traffic signal control,

    K. Othman, X. Wang, A. Shalaby, and B. Abdulhai, “Integrating multimodality and partial observability solutions into decentralized multiagent reinforcement learning adaptive traffic signal control,” IEEE Open Journal of Intelligent Transportation Systems , 2025

    2025

  17. [24]

    Hybrid choice models: Progress and challenges,

    M. Ben-Akiva, D. McFadden, K. Train, J. Walker, C. Bhat, M. Bier- laire, and M. A. Munizaga, “Hybrid choice models: Progress and challenges,” Marketing Letters, vol. 13, no. 3, pp. 163–175, 2002

    2002

  18. [25]

    Multi-agent multimodal transportation simulation for mega-cities: Application of los angeles,

    B. Y . He, Q. Jiang, H. Ma, and J. Ma, “Multi-agent multimodal transportation simulation for mega-cities: Application of los angeles,” in The 13th International Workshop on Agent-based Mobility, Traffic and Transportation Models, Methodologies and Applications , 2024

    2024

  19. [26]

    Individual mobility prediction using transit smart card data,

    Z. Zhao, H. N. Koutsopoulos, and J. Zhao, “Individual mobility prediction using transit smart card data,” Transportation Research Part C: Emerging Technologies, vol. 89, pp. 19–34, 2018

    2018

  20. [27]

    A survey on deep learning for human mobility,

    M. Luca, G. Barlacchi, B. Lepri, and L. Pappalardo, “A survey on deep learning for human mobility,” arXiv preprint arXiv:2012.02825 , 2021

    2012 arXiv

  21. [28]

    Learning universal human mobility patterns with a foundation model for cross-domain data fusion,

    H. Ma, X. Liao, Y . Liu, Q. Jiang, C. Stanford, A. Cao, and J. Ma, “Learning universal human mobility patterns with a foundation model for cross-domain data fusion,” 2024, available at SSRN

    2024

  22. [29]

    A big data architecture for digital twin creation of railway signals based on synthetic data,

    G. Salierno, L. Leonardi, and G. Cabri, “A big data architecture for digital twin creation of railway signals based on synthetic data,” IEEE Open Journal of Intelligent Transportation Systems , 2024

    2024

  23. [30]

    How people get to work at night. a discrete choice model approach towards the influence of nighttime on the choice of transport mode for commuting to work,

    J. Kapitza, “How people get to work at night. a discrete choice model approach towards the influence of nighttime on the choice of transport mode for commuting to work,” Journal of Transport Geography, 2022

    2022

  24. [31]

    Working hours and the regulations for night shift work- ers,

    T.-W. Jang, “Working hours and the regulations for night shift work- ers,” Annals of Occupational and Environmental Medicine , 2023

    2023

  25. [32]

    National household travel survey,

    F. H. Administration, “National household travel survey,” 2017

    2017

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.