Group-Aware Estimation of Elevator Trip Origin DestinationMatrices: Exact Recovery and a Cohesion Threshold
DOI:
https://doi.org/10.26713/cma.v17i3.3683Keywords:
elevator group control,, origin destination matrix, combinotorial optimizationAbstract
Recovering the trip origin-destination (OD) matrix of an elevator group from sensor data
is an ill-posed inverse problem: the boarding and alighting margins that sensors report do
not determine which origin a given alighting passenger came from. Standard estimators close
the system with a maximum-entropy (passenger-independence) prior; however, field studies
show that passengers travel in social groups, contradicting the assumption of independence.
We reformulate the problem at the group level and propose a batch-aware estimator that
assigns each group to a single primary destination, thereby matching the observed margins.
Because such a method risks being judged only on data that satisfy its own assumptions,
we study instead when it helps. We introduce a cohesion parameter ρ that controls how
strongly groups travel together and an equally informed control that separates the value of
the method from the value of the group information. We prove an exact-recovery theorem
in the cohesive limit and a closed-form law for the model-mismatch error, and show this
law implies a cohesion threshold above which the method helps and below which it does
not. Simulations, an agent-based generator in which cohesion is not imposed, and a closedloop dispatch study agree: the batch-aware estimator improves accuracy and lowers dispatch
energy and journey time when groups are sufficiently cohesive, traffic is inter-floor-dominated
and not saturated, and group sizes are sensed accurately; otherwise, a margin-only estimator
is preferable. The result is a method, its supporting theory, and a calibratable decision rule
for its use
Downloads
References
[1] B. Chen, L. Ruan, J. Ye, and J. Bian, “Elevator traffic pattern recognition based on density peak clustering,”
in Proceedings of the 2018 IEEE International Conference of Safety Produce Informatization (IICSPI), 2018,
pp. 313–316. doi:10.1109/IICSPI.2018.8690418.
[2] P. A. Markos and A. J. Dentsoras, An integrated mathematical method for traffic analysis of elevator
systems, Applied Mathematical Modelling 105 (2022), 50–80, doi:10.1016/j.apm.2021.12.027.
[3] J.-M. Kuusinen, J. Sorsa, M.-L. Siikonen and H. Ehtamo, A new model for vertical building traffic, Aalto
University (2010); related published version: A study on the arrival process of lift passengers in a multi-storey
office building, Building Services Engineering Research and Technology 33 (2012), 437–449.
[4] B. Ding et al., A hybrid approach for the analysis and prediction of elevator passenger flow in an office
building, Automation in Construction 35 (2013), 69–78, doi:10.1016/j.autcon.2013.05.005.
[5] M. Shi, X. Xu and Y. Choi, Gaussian analysis of the elevator traffic under the typical office building,
Frontiers in Computing and Intelligent Systems 8(1) (2024), 75–82, doi:10.54097/379xzj23.
[6] M. Shi, X. Y. Xu and Y. Choi, Prediction and analysis of elevator traffic flow under the LSTM neural
network, Intelligent Control and Automation 15 (2024), 63–82.
[7] C. Ciflikli and E. O. Tartan, A model for the visualization and analysis of elevator traffic, Transportation
Planning and Technology 42(8) (2019), 868–880, doi:10.1080/03081060.2019.1675326.
[8] K. Kondo, R. Ohori, K. Matsue and H. Aizu, On-the-fly mathematical formulation for estimating people
flow from elevator load data in smart building virtual sensing platforms, Journal of Computational Science
85 (2025), 102527, doi:10.1016/j.jocs.2024.102527.
[9] J.-M. Kuusinen, J. Sorsa and M.-L. Siikonen, The elevator trip origin-destination matrix estimation problem,
Transportation Science 49(3) (2015), 559–576, doi:10.1287/trsc.2013.0509.
[10] A. Malapert and J.-M. Kuusinen, Estimation of elevator passenger traffic based on the most likely elevator trip origin-destination matrices, Building Services Engineering Research and Technology 38 (2017),
doi:10.1177/0143624417707875.
[11] M.-L. Siikonen, Daily traffic profiles, Elevator World (2021); see also M.-L. Siikonen, People Flow in Buildings, Wiley, Hoboken (2021).
[12] J. Sorsa, M.-L. Siikonen, J.-M. Kuusinen and H. Hakonen, A field study and analysis of passengers arriving at lift lobbies in social groups in multi-storey office, hotel and residential buildings, Building Services
Engineering Research and Technology 42(2) (2021), doi:10.1177/0143624420983974.
[13] J.-M. Kuusinen, People Flow in Buildings – Evacuation Experiments and Modelling of Elevator Passenger
Traffic, Doctoral dissertation, Aalto University publication series Doctoral Dissertations 47/2015, Espoo,
Finland (2015).
[14] D. Gu, N. Zhang, Z. Xu, Y. Wu and Y. Tian, Urban risk assessment model to quantify earthquake-induced
elevator passenger entrapment with population heatmap, Computer-Aided Civil and Infrastructure Engineering 39(14) (2024), 2204–2222, doi:10.1111/mice.13287.
[15] C. Jia, W. Yi, Y. Wu, H. Huang, L. Zhang and L. Wu, Abnormal activity capture from passenger flow
of elevator based on unsupervised learning and fine-grained multi-label recognition, arXiv preprint (2020),
doi:10.48550/arXiv.2006.15873.
[16] C. Jia, W. Yi, Y. Wu, Z. Li, S. Zhu and L. Wu, Crowd anomaly detection based on elevator Internet
of Things technology, in: Cloud Computing, Smart Grid and Innovative Frontiers in Telecommunications,
Lecture Notes of ICST, Springer (2021), doi:10.1007/978-3-030-71061-3 1.
[17] D. Wu, S. Wu, Q. Zhao, S. Zhang, J. Qi, J. Hu and B. Lin, Computer vision-based intelligent elevator
information system for efficient demand-based operation and optimization, Journal of Building Engineering
81 (2024), 108126, doi:10.1016/j.jobe.2023.108126.
[18] J. Zhao and G. Yan, Passenger flow monitoring of elevator video based on computer vision, in: Proceedings
of the 2019 Chinese Control and Decision Conference (CCDC), Nanchang, China (2019), pp. 2089–2094,
doi:10.1109/CCDC.2019.8833248.
Published
Issue
Section
License
How to Cite
