Pieter L. van Den Berg
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Pieter L. van Den Berg has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Caroline Jeanne Jagtenberg
Department of Operations Analytics, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
5 shared publications - Ayman Abdelwahed
Mathematics & Physics Department, Faculty of Engineering, Alexandria University, Alexandria, Egypt
4 shared publications - J. Theresia van Essen
Delft Institute of Applied Mathematics, Delft University of Technology, Delft, The Netherlands
4 shared publications Karen Aardal
Delft Institute of Applied Mathematics, Delft University of Technology, Delft, The Netherlands
4 shared publications- Tobias Brandt
Department of Information System, University of Münster, Münster, Germany
4 shared publications - Jo Røislien
Department of Quality and Health Technology, Faculty of Health Sciences, University of Stavanger, Stavanger, Norway
3 shared publications
Latest Publications
- Open Access
- 18 September 2026
- Published by Elsevier in SSRN Electronic Journal
Achieving Hinterland Modal Shift Under Decentralized LSP Decision-Making: A Hinterland Planning Model for Port Authorities
Port authorities are expected to play a facilitating role for the movement of hinterland freight towards lower-emission modes, yet they do not control the shipment-level decisions of the logistics service providers (LSPs) that determine mode, route and timing. This paper develops a two-level tactical planning model that allows a port authority to evaluate policy interventions, such as environmental regulation, intermodal capacity configuration, service pricing and booking flexibility, through the decentralized operational response of multiple LSPs rather than through a centrally imposed plan. At the operational level, each LSP plans shipments over a rolling horizon, trading off transport cost, emissions and transit time under the conditions set by the tactical policies. The operational problem is formulated on a reduced network representation that substantially improves computational efficiency and makes the approach tractable for large-scale hinterland networks. The model is implemented as a decision support module within the digital twin demo developed for HAROPA Port (Le Havre-Rouen-Paris) and used for scenario-based policy evaluation. The experiments yield four managerial insights. Intermodal services can lower cost and emissions simultaneously, but time pressure shifts cargo back towards road transport. Limited intermodal capacity is the binding constraint on modal shift, but capacity expansions must be targeted rather than uniform. Pricing requires careful calibration, since both high prices and excessive subsidies distort modal choice and lead to an inefficient allocation of scarce intermodal capacity. Flexible booking improves planning efficiency, but its benefits depend strongly on coordination and diminish in fragmented markets.
- 18 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAchieving Hinterland Modal Shift Under Decentralized LSP Decision-Making: A Hinterland Planning Model for Port Authorities
Port authorities are expected to play a facilitating role for the movement of hinterland freight towards lower-emission modes, yet they do not control the shipment-level decisions of the logistics service providers (LSPs) that determine mode, route and timing. This paper develops a two-level tactical planning model that allows a port authority to evaluate policy interventions, such as environmental regulation, intermodal capacity configuration, service pricing and booking flexibility, through the decentralized operational response of multiple LSPs rather than through a centrally imposed plan. At the operational level, each LSP plans shipments over a rolling horizon, trading off transport cost, emissions and transit time under the conditions set by the tactical policies. The operational problem is formulated on a reduced network representation that substantially improves computational efficiency and makes the approach tractable for large-scale hinterland networks. The model is implemented as a decision support module within the digital twin demo developed for HAROPA Port (Le Havre-Rouen-Paris) and used for scenario-based policy evaluation. The experiments yield four managerial insights. Intermodal services can lower cost and emissions simultaneously, but time pressure shifts cargo back towards road transport. Limited intermodal capacity is the binding constraint on modal shift, but capacity expansions must be targeted rather than uniform. Pricing requires careful calibration, since both high prices and excessive subsidies distort modal choice and lead to an inefficient allocation of scarce intermodal capacity. Flexible booking improves planning efficiency, but its benefits depend strongly on coordination and diminish in fragmented markets.
- Open Access
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Design of Berth Capacity Planning Policies: Trading Off Barge Throughput and Deep-Sea Vessel Service Levels at Container Terminals
Berth capacity planning at deep-sea container terminals has historically been a one-sided problem. Terminals commit capacity to deep-sea vessels through liner service agreements, while barges, lacking contractual standing, are served on a residual and ad hoc basis. Modal split mandates and barge call planning initiatives are now making barge handling an explicit capacity planning consideration for terminals that have historically planned capacity exclusively around deep-sea vessels. We model the terminal as a hybrid queuing network in which each berth serves either one deep-sea vessel or several barges, combining class-dependent service rates with asymmetric berth occupancy, a structure absent from prior queuing models of port operations. We show that the network remains a level-independent Quasi-Birth-Death process under every static reservation policy, which admits a matrix-geometric solution and allows the entire policy space to be evaluated exactly rather than searched. This characterizes the frontier between barge throughput and deep-sea vessel waiting time across full berth dedication and partial barge acceptance limits. A Markov Decision Process, formulated as a reference rather than as an operational prescription, extends the comparison to reservations that respond to the state of the terminal. Three findings emerge. Partial reservation is dominated by either full sharing or full dedication. Dedicated barge berths exhibit diminishing throughput returns while imposing disproportionate waiting time costs. State-dependent capacity reservation can reach the same barge throughput with no berth dedicated to barges. Applied to the ECT Delta Terminal in Rotterdam, the framework maps capacity reservation decisions onto the feasible modal split frontier, giving terminal operators and port authorities a tractable basis for evaluating capacity planning decisions against modal split targets.
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDesign of Berth Capacity Planning Policies: Trading Off Barge Throughput and Deep-Sea Vessel Service Levels at Container Terminals
Berth capacity planning at deep-sea container terminals has historically been a one-sided problem. Terminals commit capacity to deep-sea vessels through liner service agreements, while barges, lacking contractual standing, are served on a residual and ad hoc basis. Modal split mandates and barge call planning initiatives are now making barge handling an explicit capacity planning consideration for terminals that have historically planned capacity exclusively around deep-sea vessels. We model the terminal as a hybrid queuing network in which each berth serves either one deep-sea vessel or several barges, combining class-dependent service rates with asymmetric berth occupancy, a structure absent from prior queuing models of port operations. We show that the network remains a level-independent Quasi-Birth-Death process under every static reservation policy, which admits a matrix-geometric solution and allows the entire policy space to be evaluated exactly rather than searched. This characterizes the frontier between barge throughput and deep-sea vessel waiting time across full berth dedication and partial barge acceptance limits. A Markov Decision Process, formulated as a reference rather than as an operational prescription, extends the comparison to reservations that respond to the state of the terminal. Three findings emerge. Partial reservation is dominated by either full sharing or full dedication. Dedicated barge berths exhibit diminishing throughput returns while imposing disproportionate waiting time costs. State-dependent capacity reservation can reach the same barge throughput with no berth dedicated to barges. Applied to the ECT Delta Terminal in Rotterdam, the framework maps capacity reservation decisions onto the feasible modal split frontier, giving terminal operators and port authorities a tractable basis for evaluating capacity planning decisions against modal split targets.
- Open Access
- 11 June 2026
- Published by Springer Nature in Health Care Management Science
- Vol. 29 (2) , 25
Ahead of the ambulance: Optimizing volunteer training
Emergency response for medical incidents is increasingly extended by community first responder (CFR) systems that dispatch nearby trained volunteers. The implementation of CFR systems has led to significant decreases in emergency response times, especially in rural areas where ambulances take longer to arrive. CFR systems that dispatch volunteers to various emergency types can increase their effectiveness by training their volunteers, enabling these volunteers to provide first aid for more emergency types. We study the problem of optimizing a CFR system’s training strategy to maximize its effectiveness given a limited budget, where the effectiveness is measured by the probability that at least one volunteer arrives before the ambulance for any given incident. We introduce an optimization model that explicitly accounts for the heterogeneous nature of volunteers’ availability and locations, as well as a solution approach that efficiently obtains optimal solutions for realistically-sized instances. We apply the optimization approach to a CFR system operating in Lincolnshire, United Kingdom. The results show that the optimization approach yields substantially larger improvements in the CFR system’s effectiveness compared to several intuitive greedy training strategies. Additionally, dispatch restrictions to limit the workload of volunteers are shown to have important implications for the optimal training strategy.
- 11 June 2026
- Published by Springer Nature in Health Care Management Science
- Vol. 29 (2) , 25
Open AccessAhead of the ambulance: Optimizing volunteer training
Emergency response for medical incidents is increasingly extended by community first responder (CFR) systems that dispatch nearby trained volunteers. The implementation of CFR systems has led to significant decreases in emergency response times, especially in rural areas where ambulances take longer to arrive. CFR systems that dispatch volunteers to various emergency types can increase their effectiveness by training their volunteers, enabling these volunteers to provide first aid for more emergency types. We study the problem of optimizing a CFR system’s training strategy to maximize its effectiveness given a limited budget, where the effectiveness is measured by the probability that at least one volunteer arrives before the ambulance for any given incident. We introduce an optimization model that explicitly accounts for the heterogeneous nature of volunteers’ availability and locations, as well as a solution approach that efficiently obtains optimal solutions for realistically-sized instances. We apply the optimization approach to a CFR system operating in Lincolnshire, United Kingdom. The results show that the optimization approach yields substantially larger improvements in the CFR system’s effectiveness compared to several intuitive greedy training strategies. Additionally, dispatch restrictions to limit the workload of volunteers are shown to have important implications for the optimal training strategy.
- 21 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
The Value of Time and Location Commitment for Decentralized Emergency Medical Services
Problem definition: Emergency medical services (EMS) in many low- and middle-income countries utilize decentralized platforms coordinating independent ambulance providers. However, significant operational challenges arise from uncertainty in provider time availability and unpredictable idle locations. These uncertainties hinder reliable service coverage and negatively impact patient outcomes. Using data from our East-African partner Rescue.co regarding the operations of their emergency platform, Flare, in Nairobi, Kenya, we investigate the relative effectiveness of enhancing provider temporal commitment (time availability) versus spatial commitment (strategic location) to improve system coverage. Methodology/results: We employ optimization models tailored to ambulance commitment uncertainty, a detailed case study, data-driven simulations, and a game-theoretic model. Our findings quantify a stark “cost of decentralization”: The coverage provided by Flare’s 340 loosely committed ambulances could potentially be matched by fewer than 15 optimally deployed fully committed units. We find that enhancing spatial commitment typically yields substantially larger coverage gains than increasing time availability alone and that both dimensions are strongly complementary. Simulations validate these effects under stochastic demand, simultaneous calls, and strategic relocation. A stylized free-entry model, in which time- and spatially committed providers enter until profits vanish, shows that spatial commitment dominates except at very low temporal availability and that adding spatially committed units typically yields higher marginal value by filling geographic gaps. Together, all three approaches highlight the significant value of spatial coordination. Managerial implications: For managers of decentralized EMS platforms, the main priority should be to enhance spatial coordination. This can be achieved by deploying a small number of platform-controlled, location-flexible units or by incentivizing providers to relocate strategically, both of which are especially effective at closing the system’s most critical coverage gaps. Because these benefits persist even when accounting for stochastic demand, simultaneous calls, and potential strategic behavior by independent providers, managers need not account for every contingency when evaluating interventions; focusing on closing persistent geographic gaps yields robust performance gains in resource-constrained settings. Funding: This research was financed in part by the Netherlands Organization for Scientific Research (NWO) in [project number VI.Veni.191E.005 and VI.Vidi.241E.016] for P.L. van den Berg, and by the Natural Sciences and Engineering Research Council of Canada (NSERC) [Grant RGPIN-2025-05848] for G. Romero. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0324 .
- 21 May 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
The Value of Time and Location Commitment for Decentralized Emergency Medical Services
Problem definition: Emergency medical services (EMS) in many low- and middle-income countries utilize decentralized platforms coordinating independent ambulance providers. However, significant operational challenges arise from uncertainty in provider time availability and unpredictable idle locations. These uncertainties hinder reliable service coverage and negatively impact patient outcomes. Using data from our East-African partner Rescue.co regarding the operations of their emergency platform, Flare, in Nairobi, Kenya, we investigate the relative effectiveness of enhancing provider temporal commitment (time availability) versus spatial commitment (strategic location) to improve system coverage. Methodology/results: We employ optimization models tailored to ambulance commitment uncertainty, a detailed case study, data-driven simulations, and a game-theoretic model. Our findings quantify a stark “cost of decentralization”: The coverage provided by Flare’s 340 loosely committed ambulances could potentially be matched by fewer than 15 optimally deployed fully committed units. We find that enhancing spatial commitment typically yields substantially larger coverage gains than increasing time availability alone and that both dimensions are strongly complementary. Simulations validate these effects under stochastic demand, simultaneous calls, and strategic relocation. A stylized free-entry model, in which time- and spatially committed providers enter until profits vanish, shows that spatial commitment dominates except at very low temporal availability and that adding spatially committed units typically yields higher marginal value by filling geographic gaps. Together, all three approaches highlight the significant value of spatial coordination. Managerial implications: For managers of decentralized EMS platforms, the main priority should be to enhance spatial coordination. This can be achieved by deploying a small number of platform-controlled, location-flexible units or by incentivizing providers to relocate strategically, both of which are especially effective at closing the system’s most critical coverage gaps. Because these benefits persist even when accounting for stochastic demand, simultaneous calls, and potential strategic behavior by independent providers, managers need not account for every contingency when evaluating interventions; focusing on closing persistent geographic gaps yields robust performance gains in resource-constrained settings. Funding: This research was financed in part by the Netherlands Organization for Scientific Research (NWO) in [project number VI.Veni.191E.005 and VI.Vidi.241E.016] for P.L. van den Berg, and by the Natural Sciences and Engineering Research Council of Canada (NSERC) [Grant RGPIN-2025-05848] for G. Romero. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0324 .
- Open Access
- 3 March 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (3) , e0343762
Simulation of phased alerting of community first responders for cardiac arrest
Background: Community First Responders (CFRs) are commonly used for out-of-hospital cardiac arrests, and advanced systems send so-called phased alerts: notifications with built-in time delays. The policy that defines these delays affects both response times and volunteer fatigue. Methods: We compare alert policies by Monte Carlo Simulation, estimating patient survival, coverage, number of alerts and redundant CFR arrivals. In the simulation, acceptance probabilities and response delays are bootstrapped from 29,307 rows of historical data covering all GoodSAM alerts in New Zealand between 1-12-2017 and 30-11-2020. We simulate distances between the patient and CFRs by assuming that CFRs are located uniformly at random in a 1-km circle around the patient, for different CFR densities. Our simulated CFRs travel with a distance-dependent speed that was estimated by linear regression on observed speeds among those responders in the above-mentioned data set that eventually reached the patient. Results: The alerting policy has a large impact on the four metrics above, and the best choice depends on volunteer density. For each volunteer density, we are able to identify a policy that improves GoodSAM New Zealand’s current policy on all four metrics. For example, when there are 30 volunteers within 1 km from the patient, sending out alerts to 7 volunteers and replacing each volunteer that rejects by a new one, is expected to save 10 additional lives per year compared to the current policy, without increasing volunteer fatigue. Our results also shed light on polices that would improve one metric while worsening another, for example, when there are 10 volunteers within 1 km from the patient, dispatching them all immediately increases our survival estimate by 11% compared to the current policy, with the downside of also increasing the redundant arrivals by 137%. Conclusions: Monte Carlo simulation can help CFR system managers identify a good policy before implementing it in practice. We recommend balancing survival and volunteer fatigue, aiming to ultimately further improve a CFR system’s effectiveness.
- 3 March 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (3) , e0343762
Open AccessSimulation of phased alerting of community first responders for cardiac arrest
Background: Community First Responders (CFRs) are commonly used for out-of-hospital cardiac arrests, and advanced systems send so-called phased alerts: notifications with built-in time delays. The policy that defines these delays affects both response times and volunteer fatigue. Methods: We compare alert policies by Monte Carlo Simulation, estimating patient survival, coverage, number of alerts and redundant CFR arrivals. In the simulation, acceptance probabilities and response delays are bootstrapped from 29,307 rows of historical data covering all GoodSAM alerts in New Zealand between 1-12-2017 and 30-11-2020. We simulate distances between the patient and CFRs by assuming that CFRs are located uniformly at random in a 1-km circle around the patient, for different CFR densities. Our simulated CFRs travel with a distance-dependent speed that was estimated by linear regression on observed speeds among those responders in the above-mentioned data set that eventually reached the patient. Results: The alerting policy has a large impact on the four metrics above, and the best choice depends on volunteer density. For each volunteer density, we are able to identify a policy that improves GoodSAM New Zealand’s current policy on all four metrics. For example, when there are 30 volunteers within 1 km from the patient, sending out alerts to 7 volunteers and replacing each volunteer that rejects by a new one, is expected to save 10 additional lives per year compared to the current policy, without increasing volunteer fatigue. Our results also shed light on polices that would improve one metric while worsening another, for example, when there are 10 volunteers within 1 km from the patient, dispatching them all immediately increases our survival estimate by 11% compared to the current policy, with the downside of also increasing the redundant arrivals by 137%. Conclusions: Monte Carlo simulation can help CFR system managers identify a good policy before implementing it in practice. We recommend balancing survival and volunteer fatigue, aiming to ultimately further improve a CFR system’s effectiveness.
- Open Access
- 18 February 2026
- Published by Elsevier in European Journal of Operational Research
- Vol. 335 (1) , 83-104
Battery replenishment and repositioning problem in electric barge networks
- 18 February 2026
- Published by Elsevier in European Journal of Operational Research
- Vol. 335 (1) , 83-104
Open AccessBattery replenishment and repositioning problem in electric barge networks
- Open Access
- 31 August 2025
- Published by Elsevier in Fire Safety Journal
- Vol. 157
Effect of fire department response time on fire damage in the Netherlands
- 31 August 2025
- Published by Elsevier in Fire Safety Journal
- Vol. 157
Open AccessEffect of fire department response time on fire damage in the Netherlands
- 1 February 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (2) , 992-1008
Modeling the Impact of Community First Responders
In community first responder (CFR) systems, traditional emergency service response is augmented by a network of trained volunteers who are dispatched via an app. A central application of such systems is out-of-hospital cardiac arrest (OHCA), where a very fast response is crucial. For a target performance level, how many volunteers are needed, and from which locations should they be recruited? We model the presence of volunteers throughout a region as a Poisson point process, which permits the computation of the response-time distribution of the first-arriving volunteer. Combining this with known survival-rate functions, we deduce survival probabilities in the cardiac arrest setting. We then use convex optimization to compute a location distribution of volunteers across the region that optimizes either the fraction of incidents with a fast response (a common measure in the industry) or patient survival in the case of OHCA. The optimal location distribution provides a bound on the best possible performance with a given number of volunteers. This can be used to determine whether introducing a CFR system in a new region is worthwhile or can serve as a guide for additional recruitment in existing systems. Effective target areas for recruitment are not always obvious because volunteers recruited from one area may be found in various areas across the city depending on the time of day; we explicitly capture this issue. We demonstrate these methods through an extended case study of Auckland, New Zealand. This paper was accepted by Carri Chan, healthcare management. Funding: This research was financed in part by the Netherlands Organization for Scientific Research (NWO) in the form of a Rubicon grant (019.172EN.016) to C. J. Jagtenberg and a Veni grant (VI.Veni.191E.005) to P. L. van den Berg. S. G. Henderson and H. Li were supported in part by National Science Foundation [Grant CMMI-2035086]. Vrije Universiteit Amsterdam and Erasmus University received funding from TKI Dinalog [Grant 2023-1-307TKI]. Some of this work has been executed under the TKI Dinalog [Grant 2023-1-307TKI]. Part of this work was completed during a research visit made possible by a [Distinguished Visitor Award] from the University of Auckland. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.04024 .
- 1 February 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71 (2) , 992-1008
Modeling the Impact of Community First Responders
In community first responder (CFR) systems, traditional emergency service response is augmented by a network of trained volunteers who are dispatched via an app. A central application of such systems is out-of-hospital cardiac arrest (OHCA), where a very fast response is crucial. For a target performance level, how many volunteers are needed, and from which locations should they be recruited? We model the presence of volunteers throughout a region as a Poisson point process, which permits the computation of the response-time distribution of the first-arriving volunteer. Combining this with known survival-rate functions, we deduce survival probabilities in the cardiac arrest setting. We then use convex optimization to compute a location distribution of volunteers across the region that optimizes either the fraction of incidents with a fast response (a common measure in the industry) or patient survival in the case of OHCA. The optimal location distribution provides a bound on the best possible performance with a given number of volunteers. This can be used to determine whether introducing a CFR system in a new region is worthwhile or can serve as a guide for additional recruitment in existing systems. Effective target areas for recruitment are not always obvious because volunteers recruited from one area may be found in various areas across the city depending on the time of day; we explicitly capture this issue. We demonstrate these methods through an extended case study of Auckland, New Zealand. This paper was accepted by Carri Chan, healthcare management. Funding: This research was financed in part by the Netherlands Organization for Scientific Research (NWO) in the form of a Rubicon grant (019.172EN.016) to C. J. Jagtenberg and a Veni grant (VI.Veni.191E.005) to P. L. van den Berg. S. G. Henderson and H. Li were supported in part by National Science Foundation [Grant CMMI-2035086]. Vrije Universiteit Amsterdam and Erasmus University received funding from TKI Dinalog [Grant 2023-1-307TKI]. Some of this work has been executed under the TKI Dinalog [Grant 2023-1-307TKI]. Part of this work was completed during a research visit made possible by a [Distinguished Visitor Award] from the University of Auckland. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.04024 .
- Open Access
Community first response for cardiac arrest: comparing phased dispatch policies through Monte Carlo simulation
Background Advanced Community First Responder (CFR) systems send so-called phased alerts: notifications with built-in time delays. The policy that defines these delays affects response times, CFR workload and the number of redundant CFR arrivals. Methods We compare policies by Monte Carlo Simulation, estimating the three metrics above. We bootstrap acceptance probabilities and response delays from 29,307 rows of historical data covering all GoodSAM alerts in New Zealand between 1-12-2017 and 30-11-2020. We simulate distances between the patient and CFRs by assuming that CFRs are located uniformly at random in a 1 km circle around the patient, for different CFR densities. Our simulated CFRs travel with a distance-dependent speed that was estimated by linear regression on observed speeds among those responders in the abovementioned data set that eventually reached the patient. Results The alerting policy has a large impact on the expected number of alerts sent, the redundant arrivals and the probability of patient survival. CFR app managers can use our results to identify a policy that displays a desirable trade-off between these performance measures.
Open AccessCommunity first response for cardiac arrest: comparing phased dispatch policies through Monte Carlo simulation
Background Advanced Community First Responder (CFR) systems send so-called phased alerts: notifications with built-in time delays. The policy that defines these delays affects response times, CFR workload and the number of redundant CFR arrivals. Methods We compare policies by Monte Carlo Simulation, estimating the three metrics above. We bootstrap acceptance probabilities and response delays from 29,307 rows of historical data covering all GoodSAM alerts in New Zealand between 1-12-2017 and 30-11-2020. We simulate distances between the patient and CFRs by assuming that CFRs are located uniformly at random in a 1 km circle around the patient, for different CFR densities. Our simulated CFRs travel with a distance-dependent speed that was estimated by linear regression on observed speeds among those responders in the abovementioned data set that eventually reached the patient. Results The alerting policy has a large impact on the expected number of alerts sent, the redundant arrivals and the probability of patient survival. CFR app managers can use our results to identify a policy that displays a desirable trade-off between these performance measures.
- 1 September 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 24 (5) , 2463-2480
Prescriptive Analytics in Urban Policing Operations
Problem definition: We consider the case of prescriptive policing, that is, the data-driven assignment of police cars to different areas of a city. We analyze key problems with respect to prediction, optimization, and evaluation as well as trade-offs between different quality measures and crime types. Academic/practical relevance: Data-driven prescriptive analytics is gaining substantial attention in operations management research, and effective policing is at the core of the operations of almost every city in the world. Given the vast amounts of data increasingly collected within smart city initiatives and the growing safety challenges faced by urban centers worldwide, our work provides novel insights on the development and evaluation of prescriptive analytics applications in an urban context. Methodology: We conduct a computational study using crime and auxiliary data on the city of San Francisco. We analyze both strong and weak prediction methods along with two optimization formulations representing the deterrence and response time impact of police vehicle allocations. We analyze trade-offs between these effects and between different crime types. Results: We find that even weaker prediction methods can produce Pareto-efficient outcomes with respect to deterrence and response time. We identify three different archetypes of combinations of prediction methods and optimization objectives that constitute the Pareto frontier among the configurations we analyze. Furthermore, optimizing for multiple crime types biases allocations in a way that generally decreases single-type performance along one outcome metric but can improve it along the other. Managerial implications: Although optimization integrating all relevant crime types is theoretically possible, it is practically challenging because each crime type requires a collectively consistent weight. We present a framework combining prediction and optimization for a subset of key crime types with exploring the impact on the remaining types to support implementation of operations-focused smart city solutions in practice.
- 1 September 2022
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Manufacturing & Service Operations Management
- Vol. 24 (5) , 2463-2480
Prescriptive Analytics in Urban Policing Operations
Problem definition: We consider the case of prescriptive policing, that is, the data-driven assignment of police cars to different areas of a city. We analyze key problems with respect to prediction, optimization, and evaluation as well as trade-offs between different quality measures and crime types. Academic/practical relevance: Data-driven prescriptive analytics is gaining substantial attention in operations management research, and effective policing is at the core of the operations of almost every city in the world. Given the vast amounts of data increasingly collected within smart city initiatives and the growing safety challenges faced by urban centers worldwide, our work provides novel insights on the development and evaluation of prescriptive analytics applications in an urban context. Methodology: We conduct a computational study using crime and auxiliary data on the city of San Francisco. We analyze both strong and weak prediction methods along with two optimization formulations representing the deterrence and response time impact of police vehicle allocations. We analyze trade-offs between these effects and between different crime types. Results: We find that even weaker prediction methods can produce Pareto-efficient outcomes with respect to deterrence and response time. We identify three different archetypes of combinations of prediction methods and optimization objectives that constitute the Pareto frontier among the configurations we analyze. Furthermore, optimizing for multiple crime types biases allocations in a way that generally decreases single-type performance along one outcome metric but can improve it along the other. Managerial implications: Although optimization integrating all relevant crime types is theoretically possible, it is practically challenging because each crime type requires a collectively consistent weight. We present a framework combining prediction and optimization for a subset of key crime types with exploring the impact on the remaining types to support implementation of operations-focused smart city solutions in practice.