SlideShare a Scribd company logo
1 of 1
Game-theoretic Patrol Strategies for Transit Systems:
the TRUSTS System and its Mobile App
Samantha Luber, Zhengyu Yin, Francesco Delle Fave, Albert Xin Jiang, Milind Tambe and John Sullivan*
University of Southern California, *Los Angeles Sheriff’s Department
The TRUSTS System [1]The Fare Evasion Problem
Fare evasion costs proof-of-payment transit systems significant losses in revenue. In
2007 alone, the Los Angeles Metro system, using proof-of-payment, suffered an
estimated revenue loss of $5.6 million due to fare evasion.
To address this shortcoming, the Los Angeles Sheriff’s Department (LASD) periodically
patrols the Metro system in order to prevent and deter fare evasion.
However, patrolling a transit system presents a number of challenges. For instance,
resource limitations prevent officers from verifying all passengers. As a consequence,
such officers periodically inspect a subset of the passengers based on a patrol strategy.
Randomization is another challenge. The complexity of a transit system makes it
impossible for human schedulers to manually produce randomized patrol strategies,
while taking into account all of the system’s scheduling constraints. As a result, the
patrol schedules often become predictable, making it easier for people to avoid buying
their ticket.
Against this background, the intelligent deployment of effective patrol strategies is a
key challenge to deter fare evasion and maximize revenue in transit systems.
Furthermore, since potential fare evaders can exploit knowledge about the patrol
strategy to avoid inspection, a randomized patrol strategy is needed for effectiveness.
MDP TRUSTS [2]
Real World Evaluation
Acknowledgements: We thank the Los Angeles Sheriff’s
Department for their exceptional collaboration.
Future Work
The Mobile Application
10h00 to 10h15
11h00 to 11h20
Officers of the LASD patrol one line of the Metro System. They perform either
train or ticket checks, depending on a large number of spatial and temporal
constraints (e.g., time of the day, trains schedule and station location as shown in
the figure). To model this problem a compact representation is used. A transition
graph is defined encoding all the possible constraints of the problem (see Figure).
The transition graph is used to define a zero-sum Bayesian Stackelberg game. One
defender (the LASD) patrols the metro line by choosing a specific path of the
transition graph. There exists then multiple types of attacker (the fare evaders),
one for each possible path. Each type can choose whether to buy his ticket or not.
The game is solved using linear programming.
Patrol schedules are derived by sampling the defender’s mixed strategy (variables
xij in the algorithm), obtained after solving the LP.
The schedules generated using TRUSTS have been tested by the LASD in several
lines of the LA Metro system. Results showed one key problem: the schedules
were not robust to the execution uncertainty related to patrolling a public transit
system.
To address this shortcoming, the TRUSTS’s schedules are transformed into plans.
The key idea is to generalize the transition graph into an MDP (see Figure below),
i.e., the outcome of each action becomes stochastic (each action might lead to a
specific state or it might end up in another one). In so doing, the element of
execution uncertainty is incorporated within the model. For instance, train or
station checks can be delayed for some time due to the officer having to deal with
some other problem.
The solution of the MDP TRUSTS linear program is a mixed strategy for the defender. A pure strategy can then be generated by sampling the mixed strategy. However, a pure strategy now does not correspond to a simple schedule anymore. Rather it
corresponds to a plan: a mapping from states to actions, where each state represents a station of a metro line and a time slot. To visualize these plans, we developed a mobile application. The application is essentially a software tool which allows each patrol
officer to visualize its schedule and to record data related to each patrol shift. As shown in the figures, the application consists of three views.
The resulting MDP is used to define a zero-sum Bayesian Stackelberg game. The
game is defined similarly to the original TRUSTS formulation. One defender (the
LASD) patrols the metro line by choosing a specific policy of the MDP. There exists
then multiple type of attacker (the fare evaders), one for each possible route. The
attacker can choose whether to buy or not his ticket. The game is solved using
linear programming (a network flow).
Schedule View: it is used to visualize the plan for
each patrol officer.
Reporting View: it is used to submit all the
information collected during a patrol shift.
Summary View: it is used to submit the
collected information to a central database
Each action specifies:
• Type of action
• Start / End time
Select a new station
Select the violation
type
Real world tests of the application have started on January 2013. Thus far, three tests have been run on the LA metro
system (on the red line). Each test was conducted together with a team of three officers of the LASD. Further tests are
planned on the blue, green and expo line and will start in the middle of May.
Thus far, both the MDP TRUSTS system and the mobile app have shown great potential to be used in several other
public transit system in the United States (e.g., in Chicago or Washington).
The use of the mobile application opens up several different possibilities for research. The key reason is that the
information collected within each shift is recorded by the reporting view. This data can then be used to improve the
quality of the schedules. For instance, data about each evader that has been caught could be used to learn its behavior.
Similarly, data about each execution interruption could be used to learn the uncertainty model related to the MDP.
References
[1] A. Jiang, Z. Yin, C. Zhang, S. Kraus, M. Tambe Game-Theoretic Randomization for Security Patrolling with
Dynamic Execution Uncertainty. In AAMAS, 2013.
[2] Yin, Z., Jiang, A. X., Johnson, M. P., Tambe, M., Kiekintveld, C., Leyton-Brown, K., Sandholm, T., and Sullivan, J. 2012.
TRUSTS: Scheduling Randomized Patrols for Fare Inspection in Transit Systems. In IAAI, 2012.

More Related Content

What's hot

Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
Samantha Luber
 
Dotnet t-drive enhancing driving directions with taxi drivers’ intelligence
Dotnet  t-drive enhancing driving directions with taxi drivers’ intelligenceDotnet  t-drive enhancing driving directions with taxi drivers’ intelligence
Dotnet t-drive enhancing driving directions with taxi drivers’ intelligence
Ecway Technologies
 
4-How share wants to practically apply the mca methodology
4-How share wants to practically apply the mca methodology4-How share wants to practically apply the mca methodology
4-How share wants to practically apply the mca methodology
SHAREPROJECT
 
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_dayKim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
Bogdan-Mihai Cîrlugea
 
Master's Project -- Andrew W. Szmurlo
Master's Project -- Andrew W. SzmurloMaster's Project -- Andrew W. Szmurlo
Master's Project -- Andrew W. Szmurlo
Andrew Szmurlo
 

What's hot (13)

Adaptive Traffic Sampling and Management Platform
Adaptive Traffic Sampling and Management PlatformAdaptive Traffic Sampling and Management Platform
Adaptive Traffic Sampling and Management Platform
 
Vehicle counting for traffic management
Vehicle counting for traffic management Vehicle counting for traffic management
Vehicle counting for traffic management
 
Urban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An OverviewUrban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An Overview
 
Traffic management system sushrut bhosale
Traffic management system sushrut bhosaleTraffic management system sushrut bhosale
Traffic management system sushrut bhosale
 
Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and i...
 
IEEE 2014 JAVA PARALLEL DISTRIBUTED PROJECTS Trajectory improves data deliver...
IEEE 2014 JAVA PARALLEL DISTRIBUTED PROJECTS Trajectory improves data deliver...IEEE 2014 JAVA PARALLEL DISTRIBUTED PROJECTS Trajectory improves data deliver...
IEEE 2014 JAVA PARALLEL DISTRIBUTED PROJECTS Trajectory improves data deliver...
 
Dotnet t-drive enhancing driving directions with taxi drivers’ intelligence
Dotnet  t-drive enhancing driving directions with taxi drivers’ intelligenceDotnet  t-drive enhancing driving directions with taxi drivers’ intelligence
Dotnet t-drive enhancing driving directions with taxi drivers’ intelligence
 
Nz Mugs Jon Slason 16 Oct2009
Nz Mugs Jon Slason 16 Oct2009Nz Mugs Jon Slason 16 Oct2009
Nz Mugs Jon Slason 16 Oct2009
 
Ite Vancourver Nelson Corridors
Ite Vancourver Nelson CorridorsIte Vancourver Nelson Corridors
Ite Vancourver Nelson Corridors
 
4-How share wants to practically apply the mca methodology
4-How share wants to practically apply the mca methodology4-How share wants to practically apply the mca methodology
4-How share wants to practically apply the mca methodology
 
Application of Cumulative Axle Model To Impute Missing Traffic Data in Defect...
Application of Cumulative Axle Model To Impute Missing Traffic Data in Defect...Application of Cumulative Axle Model To Impute Missing Traffic Data in Defect...
Application of Cumulative Axle Model To Impute Missing Traffic Data in Defect...
 
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_dayKim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
Kim-Blanco_Cirlugea_de Sherbinin_OSM_validation_Data_science_day
 
Master's Project -- Andrew W. Szmurlo
Master's Project -- Andrew W. SzmurloMaster's Project -- Andrew W. Szmurlo
Master's Project -- Andrew W. Szmurlo
 

Viewers also liked

Viewers also liked (6)

Personal Branding AAMAS
Personal Branding AAMASPersonal Branding AAMAS
Personal Branding AAMAS
 
Ancestral Causal Inference - NIPS 2016 poster
Ancestral Causal Inference - NIPS 2016 posterAncestral Causal Inference - NIPS 2016 poster
Ancestral Causal Inference - NIPS 2016 poster
 
Introduction to GSM - an Overview of Global System for Mobile Communication
Introduction to GSM - an Overview of Global System for Mobile CommunicationIntroduction to GSM - an Overview of Global System for Mobile Communication
Introduction to GSM - an Overview of Global System for Mobile Communication
 
ICML2015読み会 資料
ICML2015読み会 資料ICML2015読み会 資料
ICML2015読み会 資料
 
Gsm.....ppt
Gsm.....pptGsm.....ppt
Gsm.....ppt
 
GENERATIONS OF COMPUTER
GENERATIONS OF COMPUTERGENERATIONS OF COMPUTER
GENERATIONS OF COMPUTER
 

Similar to TRUSTS Mobile App Demo Poster (AAMAS 2013)

Density of route frequency for enforcement
Density of route frequency for enforcement Density of route frequency for enforcement
Density of route frequency for enforcement
Conference Papers
 

Similar to TRUSTS Mobile App Demo Poster (AAMAS 2013) (20)

Study of statistical models for route prediction algorithms in vanet
Study of statistical models for route prediction algorithms in vanetStudy of statistical models for route prediction algorithms in vanet
Study of statistical models for route prediction algorithms in vanet
 
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
 
Predicting Operating Train Delays into New York City using Random Forest Regr...
Predicting Operating Train Delays into New York City using Random Forest Regr...Predicting Operating Train Delays into New York City using Random Forest Regr...
Predicting Operating Train Delays into New York City using Random Forest Regr...
 
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
 
Density of route frequency for enforcement
Density of route frequency for enforcement Density of route frequency for enforcement
Density of route frequency for enforcement
 
FOLLOWING CAR ALGORITHM WITH MULTI AGENT RANDOMIZED SYSTEM
FOLLOWING CAR ALGORITHM WITH MULTI AGENT RANDOMIZED SYSTEMFOLLOWING CAR ALGORITHM WITH MULTI AGENT RANDOMIZED SYSTEM
FOLLOWING CAR ALGORITHM WITH MULTI AGENT RANDOMIZED SYSTEM
 
Neural Network Based Parking via Google Map Guidance
Neural Network Based Parking via Google Map GuidanceNeural Network Based Parking via Google Map Guidance
Neural Network Based Parking via Google Map Guidance
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment Algorithms
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment Algorithms
 
Travel time prediction using svm and wma
Travel time prediction using svm and wmaTravel time prediction using svm and wma
Travel time prediction using svm and wma
 
Review of Fuzzy Model
Review of Fuzzy Model Review of Fuzzy Model
Review of Fuzzy Model
 
Fuzzy Model Presentation
Fuzzy Model Presentation Fuzzy Model Presentation
Fuzzy Model Presentation
 
Travel Time Prediction Using Dedicated Short-Range Communications Probe Data
Travel Time Prediction Using Dedicated Short-Range Communications Probe DataTravel Time Prediction Using Dedicated Short-Range Communications Probe Data
Travel Time Prediction Using Dedicated Short-Range Communications Probe Data
 
T drive enhancing driving directions with taxi drivers’ intelligence
T drive enhancing driving directions with taxi drivers’ intelligenceT drive enhancing driving directions with taxi drivers’ intelligence
T drive enhancing driving directions with taxi drivers’ intelligence
 
With saloni in ijarcsse
With saloni in ijarcsseWith saloni in ijarcsse
With saloni in ijarcsse
 
Improving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBSImproving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBS
 
T drive enhancing driving directions with taxi drivers’ intelligence
T drive enhancing driving directions with taxi drivers’ intelligenceT drive enhancing driving directions with taxi drivers’ intelligence
T drive enhancing driving directions with taxi drivers’ intelligence
 
JAVA 2013 IEEE DATAMINING PROJECT T drive enhancing driving directions with t...
JAVA 2013 IEEE DATAMINING PROJECT T drive enhancing driving directions with t...JAVA 2013 IEEE DATAMINING PROJECT T drive enhancing driving directions with t...
JAVA 2013 IEEE DATAMINING PROJECT T drive enhancing driving directions with t...
 
A Simulation-Based Dynamic Traffic Assignment Model With Combined Modes
A Simulation-Based Dynamic Traffic Assignment Model With Combined ModesA Simulation-Based Dynamic Traffic Assignment Model With Combined Modes
A Simulation-Based Dynamic Traffic Assignment Model With Combined Modes
 
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic FlowUsing Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
 

More from Samantha Luber

Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
Samantha Luber
 

More from Samantha Luber (20)

Media-based Querying and Searching
Media-based Querying and SearchingMedia-based Querying and Searching
Media-based Querying and Searching
 
User Prompts for TRUSTS Mobile App Demonstration (AAMAS 2013)
User Prompts for TRUSTS Mobile App Demonstration (AAMAS 2013)User Prompts for TRUSTS Mobile App Demonstration (AAMAS 2013)
User Prompts for TRUSTS Mobile App Demonstration (AAMAS 2013)
 
Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
Game-theoretic Patrol Strategies for Transit Systems (Slideshow deck)
 
Autonomous Robot Band Presentation
Autonomous Robot Band PresentationAutonomous Robot Band Presentation
Autonomous Robot Band Presentation
 
Object-retrieving Autonomous Robotic Arm
Object-retrieving Autonomous Robotic ArmObject-retrieving Autonomous Robotic Arm
Object-retrieving Autonomous Robotic Arm
 
Web-controlled Car Poster
Web-controlled Car PosterWeb-controlled Car Poster
Web-controlled Car Poster
 
Autonomous Band Project Writeup
Autonomous Band Project WriteupAutonomous Band Project Writeup
Autonomous Band Project Writeup
 
Electronic Dance Music Presentation
Electronic Dance Music PresentationElectronic Dance Music Presentation
Electronic Dance Music Presentation
 
Digital Tuner Project Final Report
Digital Tuner Project Final ReportDigital Tuner Project Final Report
Digital Tuner Project Final Report
 
Digital Tuner Project Final Presentation
Digital Tuner Project Final PresentationDigital Tuner Project Final Presentation
Digital Tuner Project Final Presentation
 
Strategic Trading in Credit Networks
Strategic Trading in Credit NetworksStrategic Trading in Credit Networks
Strategic Trading in Credit Networks
 
Phi Sigma Rho Engineering Sorority
Phi Sigma Rho Engineering SororityPhi Sigma Rho Engineering Sorority
Phi Sigma Rho Engineering Sorority
 
Efficient Belief Propagation in Depth Finding
Efficient Belief Propagation in Depth FindingEfficient Belief Propagation in Depth Finding
Efficient Belief Propagation in Depth Finding
 
Gangs and Violence in Brazil
Gangs and Violence in BrazilGangs and Violence in Brazil
Gangs and Violence in Brazil
 
MSAIL Mass Meeting Winer 2011
MSAIL Mass Meeting Winer 2011MSAIL Mass Meeting Winer 2011
MSAIL Mass Meeting Winer 2011
 
Cognitive Science Artificial Intelligence
Cognitive Science Artificial IntelligenceCognitive Science Artificial Intelligence
Cognitive Science Artificial Intelligence
 
AbioCor Heart System
AbioCor Heart SystemAbioCor Heart System
AbioCor Heart System
 
The AbioCor System: Overview
The AbioCor System: OverviewThe AbioCor System: Overview
The AbioCor System: Overview
 
Spinal Disc Implants
Spinal Disc ImplantsSpinal Disc Implants
Spinal Disc Implants
 
SCAI Presentation
SCAI PresentationSCAI Presentation
SCAI Presentation
 

Recently uploaded

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
vu2urc
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 

Recently uploaded (20)

GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdf
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 

TRUSTS Mobile App Demo Poster (AAMAS 2013)

  • 1. Game-theoretic Patrol Strategies for Transit Systems: the TRUSTS System and its Mobile App Samantha Luber, Zhengyu Yin, Francesco Delle Fave, Albert Xin Jiang, Milind Tambe and John Sullivan* University of Southern California, *Los Angeles Sheriff’s Department The TRUSTS System [1]The Fare Evasion Problem Fare evasion costs proof-of-payment transit systems significant losses in revenue. In 2007 alone, the Los Angeles Metro system, using proof-of-payment, suffered an estimated revenue loss of $5.6 million due to fare evasion. To address this shortcoming, the Los Angeles Sheriff’s Department (LASD) periodically patrols the Metro system in order to prevent and deter fare evasion. However, patrolling a transit system presents a number of challenges. For instance, resource limitations prevent officers from verifying all passengers. As a consequence, such officers periodically inspect a subset of the passengers based on a patrol strategy. Randomization is another challenge. The complexity of a transit system makes it impossible for human schedulers to manually produce randomized patrol strategies, while taking into account all of the system’s scheduling constraints. As a result, the patrol schedules often become predictable, making it easier for people to avoid buying their ticket. Against this background, the intelligent deployment of effective patrol strategies is a key challenge to deter fare evasion and maximize revenue in transit systems. Furthermore, since potential fare evaders can exploit knowledge about the patrol strategy to avoid inspection, a randomized patrol strategy is needed for effectiveness. MDP TRUSTS [2] Real World Evaluation Acknowledgements: We thank the Los Angeles Sheriff’s Department for their exceptional collaboration. Future Work The Mobile Application 10h00 to 10h15 11h00 to 11h20 Officers of the LASD patrol one line of the Metro System. They perform either train or ticket checks, depending on a large number of spatial and temporal constraints (e.g., time of the day, trains schedule and station location as shown in the figure). To model this problem a compact representation is used. A transition graph is defined encoding all the possible constraints of the problem (see Figure). The transition graph is used to define a zero-sum Bayesian Stackelberg game. One defender (the LASD) patrols the metro line by choosing a specific path of the transition graph. There exists then multiple types of attacker (the fare evaders), one for each possible path. Each type can choose whether to buy his ticket or not. The game is solved using linear programming. Patrol schedules are derived by sampling the defender’s mixed strategy (variables xij in the algorithm), obtained after solving the LP. The schedules generated using TRUSTS have been tested by the LASD in several lines of the LA Metro system. Results showed one key problem: the schedules were not robust to the execution uncertainty related to patrolling a public transit system. To address this shortcoming, the TRUSTS’s schedules are transformed into plans. The key idea is to generalize the transition graph into an MDP (see Figure below), i.e., the outcome of each action becomes stochastic (each action might lead to a specific state or it might end up in another one). In so doing, the element of execution uncertainty is incorporated within the model. For instance, train or station checks can be delayed for some time due to the officer having to deal with some other problem. The solution of the MDP TRUSTS linear program is a mixed strategy for the defender. A pure strategy can then be generated by sampling the mixed strategy. However, a pure strategy now does not correspond to a simple schedule anymore. Rather it corresponds to a plan: a mapping from states to actions, where each state represents a station of a metro line and a time slot. To visualize these plans, we developed a mobile application. The application is essentially a software tool which allows each patrol officer to visualize its schedule and to record data related to each patrol shift. As shown in the figures, the application consists of three views. The resulting MDP is used to define a zero-sum Bayesian Stackelberg game. The game is defined similarly to the original TRUSTS formulation. One defender (the LASD) patrols the metro line by choosing a specific policy of the MDP. There exists then multiple type of attacker (the fare evaders), one for each possible route. The attacker can choose whether to buy or not his ticket. The game is solved using linear programming (a network flow). Schedule View: it is used to visualize the plan for each patrol officer. Reporting View: it is used to submit all the information collected during a patrol shift. Summary View: it is used to submit the collected information to a central database Each action specifies: • Type of action • Start / End time Select a new station Select the violation type Real world tests of the application have started on January 2013. Thus far, three tests have been run on the LA metro system (on the red line). Each test was conducted together with a team of three officers of the LASD. Further tests are planned on the blue, green and expo line and will start in the middle of May. Thus far, both the MDP TRUSTS system and the mobile app have shown great potential to be used in several other public transit system in the United States (e.g., in Chicago or Washington). The use of the mobile application opens up several different possibilities for research. The key reason is that the information collected within each shift is recorded by the reporting view. This data can then be used to improve the quality of the schedules. For instance, data about each evader that has been caught could be used to learn its behavior. Similarly, data about each execution interruption could be used to learn the uncertainty model related to the MDP. References [1] A. Jiang, Z. Yin, C. Zhang, S. Kraus, M. Tambe Game-Theoretic Randomization for Security Patrolling with Dynamic Execution Uncertainty. In AAMAS, 2013. [2] Yin, Z., Jiang, A. X., Johnson, M. P., Tambe, M., Kiekintveld, C., Leyton-Brown, K., Sandholm, T., and Sullivan, J. 2012. TRUSTS: Scheduling Randomized Patrols for Fare Inspection in Transit Systems. In IAAI, 2012.