Pith. sign in

REVIEW 3 major objections 6 minor 22 references

Robust Management of Airport Security Queues Considering Passenger Non-compliance with Chance-Constrained Optimization

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read A chance-constrained arrival-slot policy cuts airport security waiting time by 85 percent versus no control, and its reduction is 30 percent larger than a full-compliance policy's.

desk verdict Plausible operational idea let down by a simulation that contradicts its own equation and a chance-constraint reformulation missing a square root—the headline TTS numbers should not be trusted as written. read the letter →

arxiv 2505.05717 v1 pith:DXK7C5TF submitted 2025-05-09 eess.SY cs.SY

classification eess.SYcs.SY
keywords airportsecurityqueueschance-constrainedoptimizationpassengernon-compliancearrivaltimeslotrecommendationsecond-orderconeprogramtotalsavingsBarcelona-ElPratqueuingmodel
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

The paper proposes to manage airport security queues by recommending each passenger an arrival time slot, and argues that the recommendation should be computed under uncertainty about whether passengers will follow it. It formulates the assignment as a chance-constrained optimization problem: with probability at least $1-\gamma$, the number of passengers arriving in each 15-minute slot must stay below the security checkpoint's capacity. On a sample day of 260 departing flights and 49,034 seats at Barcelona-El Prat Terminal 1, the resulting policy cuts total security waiting time by about 85 percent relative to no control, while a deterministic policy that assumes full compliance cuts it by 65 percent. The paper takes the gap as evidence that ignoring compliance variability understates the value of arrival-slot scheduling.

What carries the argument

The load-bearing object is a second-order cone program with chance constraints. For each flight $i$ and time slot $t$, the decision variable $x_{it}$ is the fraction of the flight's passengers recommended to arrive in slot $t$, and a random compliance rate $\alpha_{it} \sim N(\mu_{it}, \sigma_{it})$ determines how many actually follow the recommendation. Passengers who do not comply are assumed to arrive according to a fixed lead-time distribution from a 2013 airport operations study. The realized number of passengers arriving in each slot is written through matrices that aggregate compliant and non-compliant arrivals, and the capacity constraint is imposed as $P(\text{arrivals} \le C_t) \ge 1-\gamma$, which is reformulated into a second-order cone inequality. A simulation algorithm samples compliance rates, generates realized arrival times, and computes total time savings in passenger-hours from the resulting queuing curves.

What would settle it

Compare Algorithm 1's simulated arrival curve for the Barcelona sample day with observed security-checkpoint arrival times on that day: if the realized distribution of non-compliant passengers departs materially from the 2013 lead-time distribution, or if measured total waiting time reduction does not approach 85 percent, the central claim fails. A less costly check is to re-estimate the lead-time distribution from Barcelona checkpoint data and rerun the same optimization and simulation.

Watch

Extended reading notes

Core claim

The central claim is that a chance-constrained reformulation of arrival-slot assignment yields a security queue policy that beats both first-come-first-served queuing and deterministic full-compliance scheduling. On the Barcelona sample day, the chance-constrained policy produces 8,161 passenger-hours of total time savings, an 85 percent decrease in total waiting time, compared with 6,009 passenger-hours and a 65 percent decrease for the deterministic policy. The chance-constrained solution spreads passengers from the same flight across several neighboring slots, especially during morning and midday peaks, while the deterministic solution assigns most flights a single slot. The paper also reports that, in the no-control baseline, some passengers miss their flights, whereas neither optimized policy causes missed flights. It identifies the mechanism as the spreading of arrivals induced by modeling compliance variance: queues do not build up even though the objective, which penalizes deviation from passengers' preferred arrival time, is higher than in the deterministic case.

Load-bearing premise

The model assumes that passengers who ignore a recommended arrival slot arrive exactly according to a fixed 2013 lead-time distribution, independent of what they were told; if receiving a recommendation changes their timing, or if Barcelona passengers in 2023 have a different lead-time pattern, the simulated arrival curves and the 85 percent savings would change materially.

Editorial extensions

If this is right

  • If an airport adopted the chance-constrained arrival-slot policy, security waiting time on a day shaped like the Barcelona sample would drop by about 85 percent, and no passengers would miss flights in the simulation.
  • The 30-percent-larger reduction over deterministic full-compliance scheduling implies that planning as though passengers will follow recommendations leaves real queue reductions on the table.
  • The policy spreads each flight's passengers over neighboring slots in peak periods, which keeps slot arrivals under capacity without forcing everyone to arrive much earlier.
  • Sensitivity analysis shows the reliability factor $\gamma$, the mean compliance rate, and the compliance standard deviation can be tuned: stricter robustness or higher variance pushes recommendations earlier, while higher compliance reduces the need for early arrival.
  • The same chance-constrained model can be rerun with re-estimated compliance parameters to adapt recommendations over time.

Reading between the lines

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

  • An editorial extension: the simulation treats non-compliant passengers as arriving exactly according to the fixed lead-time distribution, as if the recommendation had not been made; a more realistic model would let the recommendation itself shift non-compliant arrivals, and that could change the 85 percent figure.
  • Another extension: the 85 percent figure measures security-queue waiting only; a complete welfare comparison would subtract the added airport time passengers spend because recommendations move them earlier, which the objective already penalizes but the headline metric does not.
  • The assumption of one passenger per seat (100 percent load factor) means real flights with empty seats and no-shows would produce lower realized arrivals, so the policy's safety margins could perhaps be relaxed.
  • The same chance-constrained assignment logic could be transferred to other appointment systems with uncertain adherence, such as baggage drop, border control, or vaccination queues.
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

3 major / 6 minor

Summary. The paper proposes a chance-constrained second-order cone program to assign passengers to recommended arrival time slots at airport security, explicitly modeling uncertain passenger non-compliance. The model is tested on a sample day of Barcelona airport flights, and the authors report an 85% reduction in total security waiting time relative to no control, and a 30% larger reduction than a deterministic full-compliance policy. A sensitivity analysis over the reliability factor and compliance parameters is also presented.

Significance. If the model were correctly formulated and the numerical results were reproducible, the paper would make a useful contribution to airport queue management by integrating compliance variability into an optimization framework. The use of a real flight schedule and the parametric sensitivity analysis are strengths. However, the chance-constraint reformulation contains a serious mathematical error, and the simulation algorithm used to generate the headline TTS figures is inconsistent with the stated model. In addition, the compliance parameters are assumed, not calibrated, so the paper's central numerical claims are not currently supported. The contribution therefore cannot be assessed until these issues are addressed.

major comments (3)
  1. [Section III-C, Eqs. (17)-(18)] The reformulation of the chance constraint is algebraically incorrect. For ξ_t ~ N(μ_t, Σ_t), the standardized quantity is (C_t - μ_t^T \tilde{D} A_t x_t) / sqrt(x_t^T A_t^T \tilde{D} Σ_t \tilde{D} A_t x_t), which follows a standard normal distribution. The denominator in Eq. (17) and the norm in Eq. (18) should be the square root of the quadratic form, not the norm of the quadratic form itself. As written, the SOCP does not enforce the intended probability bound, so the reliability guarantee claimed for the robust policy is not established.
  2. [Section IV, Algorithm 1] The simulation of non-compliant passengers uses x_it \hat{α}_it β_is, whereas Eq. (8) requires x_it (1−\hat{α}_it) β_is. Since \hat{α}_it ≈ 0.7 in the base case, the total probability mass assigned by the algorithm is approximately 2 Σ_t x_it \hat{α}_it, which exceeds 1 and is not normalized. Consequently, the realized arrival distributions—and the reported TTS values of 8,161 and 6,009 passenger-hours (an 85% and 65% reduction)—are not generated by the model described in Eq. (8). This invalidates the paper's headline numerical claims.
  3. [Section IV and Figure 2] The compliance parameters μ_it and σ_it are set to arbitrary values (α=0.7, σ=0.2 in the base case) without calibration or a demonstrably realistic range. Because the TTS is computed from simulation and the optimization objective is passenger cost rather than waiting time, the reported 85% reduction is an empirical claim dependent on these assumed parameters. The paper does not show that the 30% improvement relative to the deterministic policy is robust to reasonable variations in compliance behavior, so this central comparison is not substantiated.
minor comments (6)
  1. [Eq. (8)] The summation index t is used both for the arrival slot and for the recommendation slot; using t′ for the recommendation slot would avoid ambiguity.
  2. [Eq. (11)] The term 'multinomial normal distribution' should be 'multivariate normal distribution'.
  3. [Eq. (17)] The expression \tilde{D} ξ_t^T A_t x_t has a dimension mismatch; the matrix \tilde{D} appears to be applied to the row vector ξ_t^T. It should presumably be ξ_t^T \tilde{D} A_t x_t.
  4. [Eq. (20)] The right-hand side uses D instead of \tilde{D}; this appears to be a typo.
  5. [Section IV] The phrase '30% increase in the reduction' should be clarified as a relative increase (85% versus 65% reduction) to avoid confusion with a 30-percentage-point change.
  6. [Algorithm 1] The pseudocode does not specify normalization of \hat{p}_i; if the total mass is not 1, the sampling step is undefined.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the chance-constrained policy, TTS figures, and 85%/30% comparisons are obtained from an external lead-time distribution and simulation, not from fitted inputs or self-citations.

full rationale

The paper's derivation chain is not circular. The lead-time distribution is taken from an external ACRP 2013 report; the compliance parameters (mean 0.7, std 0.2) and reliability factor gamma=0.01 are assumed inputs rather than calibrated to hit the headline TTS values. The optimization objective is a stated quadratic/linear cost on recommendation deviation, not a fitted surrogate for total waiting time, and the total time savings are computed by simulating passenger arrivals (Algorithm 1) against a queuing model and comparing with the no-control baseline. No central claim is justified by a self-citation chain or by a uniqueness theorem from the authors' prior work, and no prediction quantity is defined in terms of the result it is said to predict. A separate concern, outside circularity, is that Algorithm 1's non-compliance term appears to use alpha-hat where Eq. (8) specifies (1-alpha), and the sampled distribution is not renormalized; if so, the reported 8,161 and 6,009 passenger-hour TTS values may not be the outputs of the stated model. That is an implementation/consistency risk, not a case of the derivation reducing to its own inputs.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central results depend on several chosen parameters and external data assumptions. The compliance distribution parameters, the reliability factor, the capacity, and the recommendation window are all set by hand without empirical calibration. The behavioral and queueing assumptions are plausible but unverified. No new physical entities are introduced, but the statistical model of compliance is an idealized construct that the results hinge on.

free parameters (5)
  • Mean compliance rate µ_it = 0.7
    Chosen uniformly for all flights and slots; no empirical estimate is provided. The headline TTS values depend directly on this assumption.
  • Compliance standard deviation σ_it = 0.2
    Chosen to represent day-to-day and flight-to-flight variability; no data support it, and it directly affects the spread of the chance constraint.
  • Reliability factor γ = 0.01
    Chosen as the allowable probability of capacity violation; sensitivity analysis varies it but the primary results use 0.01.
  • Security capacity C_t = 800 passengers per 15-min slot
    Assumed value; the paper does not cite a capacity figure for Barcelona Terminal 1, and the 85% reduction is calibrated to this value.
  • Recommendation window M_i = 16 slots (4 hours)
    Chosen because the support of the lead-time distribution is contained in the 4 hours before departure; this constrains the feasible set.
assumptions (6)
  • domain assumption Passenger compliance rate α_it follows a normal distribution N(µ_it, σ_it) for every flight and slot.
    Eq (7). A normal distribution on a probability can exceed [0,1] with positive probability, and no truncation is applied. The covariance structure across flights and slots is never specified.
  • domain assumption Non-compliant passengers arrive at security according to the ACRP 2013 lead-time distribution, independent of their recommended slot.
    Eq (8) and Algorithm 1. The lead-time distribution is a skew normal with mean 64 minutes, standard deviation 30, skewness 3, taken from an external report and not validated against Barcelona 2023 data.
  • ad hoc to paper The random vector ξ_t is jointly Gaussian with known covariance Σ_t.
    Eq (15). The SOC reformulation requires this, but Σ_t is never defined, and the vector ξ_t as defined does not contain the compliance rates for all recommendation slots that enter Eq (8)-(9).
  • domain assumption The security system behaves as an FCFS queue with a deterministic service rate equal to the slot capacity C_t.
    Section III-A. Real security throughput varies with staff levels, passenger mix, and screening procedures; the model assumes a fixed 800-per-15-min service rate.
  • domain assumption Every seat is occupied, so the passenger count on flight i equals the number of seats d_i.
    Section III-C states the 100% load factor assumption. Real load factors are typically below 100%, which changes the scale of arrivals and the TTS.
  • ad hoc to paper The cost function in Eq (6), quadratic before the mean lead time and linear after, represents passenger disutility from the recommended arrival time.
    Section III-B. The authors acknowledge this is an assumption; it affects which schedule is selected, though the TTS is computed separately from the objective.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Management of Airport Security Queues Considering Passenger Non-compliance with Chance-Constrained Optimization." pith.science (2026). https://pith.science/paper/DXK7C5TF

@misc{pith2026250505717,
  author       = {Pith},
  title        = {Pith review of: Robust Management of Airport Security Queues Considering Passenger Non-compliance with Chance-Constrained Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DXK7C5TF}},
  note         = {Machine review of arXiv:2505.05717}
}
read the original abstract

The long waiting time at airport security has become an emergent issue as demand for air travel continues to grow. Not only does queuing at security cause passengers to miss their flights, but also reduce the amount of time passengers spend at the airport post-security, potentially leading to less revenue for the airport operator. One of the key issues to address to reduce waiting time is the management of arrival priority. As passengers on later flights can arrive before passengers on earlier flights, the security system does not always process passengers in the order of the degree of urgency. In this paper, we propose a chance-constrained optimization model that decides in which time slot passengers should be recommended to arrive. We use chance constraints to obtain solutions that take the uncertainty in passenger non-compliance into account. The experimental results, based on a sample day of flight schedules at the Barcelona airport, show a reduction of 85% in the total waiting time. Compared to the deterministic case, in which passengers are assumed to fully comply with the recommendations, we see a 30% increase in the reduction of the total waiting time. This highlights the importance of considering variation in passenger compliance in the management of airport security queues.

Figures

Figures reproduced from arXiv: 2505.05717 by the authors.

Figure 1
Figure 1. It is unlikely that passengers choose to arrive more [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Solution to the two optimization problems. A) So [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Queuing diagram of security checkpoint. We assume a capacity of 800 passengers per time slot [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 1
Figure 1. Figure 1: Algorithm 1 Generating Passenger Arrivals at Security a = [] ▷ Initialize an empty list of passenger arrival time for i ∈ F do Initialize pˆi = [] ▷ Initialize an empty array for the probability mass of arrival time of passengers on flight i under a control policy for …
Figure 4
Figure 4. Figure 4: Sensitivity analysis. We vary one parameter while keeping the rest fixed for comparison. (A)-(C) shows the solutions [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 20 canonical work pages

  1. [1]

    Cody, ”Airport Wait Times 2024”, Bounce, Oct., 2024, https: //bounce.com/blog/airport-wait-times-2024

    C. Cody, ”Airport Wait Times 2024”, Bounce, Oct., 2024, https: //bounce.com/blog/airport-wait-times-2024

  2. [2]

    Bertsimas and M

    D. Bertsimas and M. Sim, ”The Price of Robustness,” Operations Research , vol. 52, no. 1, pp. 35-53, 2004. https://doi.org/10.1287/opre.1030.0065

  3. [3]

    Alexis, F

    B. Alexis, F. Eric, A. Sameer, D. Daniel, ”Schedule Optimization and Staff Allocation for Airport Security Checkpoints using Guided Simulated Annealing and Integer Linear Programming”, Journal of Air Transport Management, vol. 124, 2025

  4. [4]

    Perez, L

    E. Perez, L. Taunton, and J. A. Sefair, ”A Simulation- Optimization Approach to Improve the Allocation of Security Screening Resources in Airport Terminal Checkpoints,” 2021 Winter Simulation Conference (WSC) , pp. 1–11, Dec. 2021. https://doi.org/10.1109/WSC52266.2021.9715421

  5. [5]

    Ronald, ”An application of queuing theory to airport pas- senger security”, Interfaces, vol

    G. Ronald, ”An application of queuing theory to airport pas- senger security”, Interfaces, vol. 9, no. 4, pp. 117-123, 1979. https://doi.org/10.1287/inte.9.4.117

  6. [6]

    International Air Transportation Association, ”Global Air Passenger Reaches Record High in 2024”, Jan., 2025, https://www.iata.org/en/ pressroom/2025-releases/2025-01-30-01/

  7. [7]

    J. Chen, L. Chen, and D. Sun, ”Air traffic flow management under uncertainty using chance-constrained optimization”, Transportation Re- search Part B: Methodological , vol. 102, pp. 124-141, 2017

  8. [8]

    Mun Hwan, P

    K. Mun Hwan, P. Jin Woo, and C. Yu Jin, ”A Study on the Effects of Waiting Time for Airport Security Screening Service on Passengers’ Emotional Responses and Airport Image”, Sustainability, vol. 12, no. 24, pp. 10634-10640, 2020. https://doi.org/10.3390/su122410634

Show all 22 references
  1. [9]

    Patrick, H

    K. Patrick, H. Robert, F. Eric, and H. Belinda, ”Guidelines for Preparing Peak Period and Operational Profiles (2013),” Na- tional Academies of Sciences, Engineering, and Medicine, 2013. https://doi.org/10.17226/22647

  2. [10]

    Robert, S

    L. Robert, S. Ilya, and R. Bo, ”Virtual queuing at airport security lanes”, European Jorunal of Operations Research, vol. 225, no. 1, pp. 153-165, 2013

  3. [11]

    Daniel, ”Here are the airports with the longest security wait times during the holidays: Study”, fox news, Nov., 2024, https://www

    M. Daniel, ”Here are the airports with the longest security wait times during the holidays: Study”, fox news, Nov., 2024, https://www. livenowfox.com/news/airports-longest-security-wait-times-holidays

  4. [12]

    Zachary, M

    M. Zachary, M. John, G. Adam, P. Caleb, and D. Luigi, ”Expediting airport security queues through advance lane assignment”, J Transp Secur, vol. 15, issue 3-4, pp. 245-262, 2022

  5. [13]

    Qsensor, The 11 Worst European Airports for Security Queues - June Edition, Qsensor, Jun., 2023

  6. [14]

    Burkard, M

    R. Burkard, M. Dell’Amico, and S. Martello, Assignment Problems . Society for Industrial and Applied Mathematics, 2009

  7. [15]

    Alodhaibi, R

    S. Alodhaibi, R. L. Burdett, and P. K. D. V . Yarlagadda, ”Im- pact of passenger-arrival patterns in outbound processes of air- ports,” Procedia Manufacturing , vol. 30, pp. 323-330, 2019. https://doi.org/10.1016/j.promfg.2019.02.046

  8. [16]

    Gary, ”One of Every Seven Travelers Miss Their Flights Because of Long Airport Security Lines”, Forbes News, Jun., 2018

    S. Gary, ”One of Every Seven Travelers Miss Their Flights Because of Long Airport Security Lines”, Forbes News, Jun., 2018

  9. [17]

    Geoffrey, M.M

    S. Geoffrey, M.M. Miguel, D. Daniel, and M. Catherine, ”Simulation- Optimisation-Based Decision Support System for Managing Airport Security Resources”, EUROSIM 2023, pp. 140-145, 2024, https://doi. org/10.1007/978-3-031-68438-8 11

  10. [18]

    https://crp.trb.org/acrp0715/ wp-content/themes/acrp-child/documents/029/original/ACRP 23 Airport Passenger-Related Processing Rates Guidebook.pdf

    Transportation Research Board, ”Airport Passenger-Related Processing Rates Guidebook,” Airport Cooperative Research Program (ACRP) Report 23, 2009. https://crp.trb.org/acrp0715/ wp-content/themes/acrp-child/documents/029/original/ACRP 23 Airport Passenger-Related Processing Ra...

  11. [19]

    https://crp.trb.org/acrp0715/wp-content/themes/acrp-child/documents/ 186/original/acrp wo22.pdf

    Transportation Research Board, ”Passenger Value of Time, Benefit-Cost Analysis and Airport Capital Investment Decisions,” Airport Cooperative Research Program (ACRP) Report 107, 2015. https://crp.trb.org/acrp0715/wp-content/themes/acrp-child/documents/ 186/original/acrp wo22.pdf

  12. [20]

    Ruiting, K

    W. Ruiting, K. Patrick, Z. Teng, S. Jairo, V . Aashrith, B. Hoseinali, and M. Scott, ”Robust routing for a mixed fleet of heavy-duty trucks with pickup and delivery under energy consumption uncertainty”, Applied Energy, vol. 368, 2024

  13. [21]

    Z. A. Marshall, J. H. Mott, A. J. Gottwald, C. A. Patrick, and L. R. Dy, ”Expediting airport security queues through advanced lane assignment,” Journal of Transportation Security , vol. 15, pp. 245-262,

  14. [2022]

    https://doi.org/10.1007/s12198-022-00247-9

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.