Skip to main content

Shih-Fen Cheng

Singapore Management University, School of Information Systems, Faculty Member

Followers

12

Following

1

Co-author

1

Public Views

University of Illinois at Urbana-Champaign

University of Delaware

Université du Luxembourg

Chandini Ramesh

InterestsView All (6)

Uploads

Papers by Shih-Fen Cheng

Multiscale Generative Models: Improving Performance of a Generative Model Using Feedback from Other Dependent Generative Models

Proceedings of the AAAI Conference on Artificial Intelligence

Realistic fine-grained multi-agent simulation of real-world complex systems is crucial for many d... more Realistic fine-grained multi-agent simulation of real-world complex systems is crucial for many downstream tasks such as reinforcement learning. Recent work has used generative models (GANs in particular) for providing high-fidelity simulation of real-world systems. However, such generative models are often monolithic and miss out on modeling the interaction in multi-agent systems. In this work, we take a first step towards building multiple interacting generative models (GANs) that reflects the interaction in real world. We build and analyze a hierarchical set-up where a higher-level GAN is conditioned on the output of multiple lower-level GANs. We present a technique of using feedback from the higher-level GAN to improve performance of lower-level GANs. We mathematically characterize the conditions under which our technique is impactful, including understanding the transfer learning nature of our set-up. We present three distinct experiments on synthetic data, time series data, an...

Multi-Agent Task Assignment for Mobile Crowdsourcing under Trajectory Uncertainties

Adaptive Agents and Multi-Agents Systems, May 4, 2015

In this work, we investigate the problem of mobile crowdsourcing, where workers are financially m... more In this work, we investigate the problem of mobile crowdsourcing, where workers are financially motivated to perform location-based tasks physically. Unlike current industry practice that relies on workers to manually browse and filter tasks to perform, we intend to automatically make task recommendations based on workers' historical trajectories and desired time budgets. However, predicting workers' trajectories is inevitably faced with uncertainties, as no one will take exactly the same route every day; yet such uncertainties are oftentimes abstracted away in the known literature. In this work, we depart from the deterministic modeling and study the stochastic task recommendation problem where each worker is associated with several predicted routine routes with probabilities. We formulate this problem as a stochastic integer linear program whose goal is to maximize the expected total utility achieved by all workers. We further exploit the separable structure of the formulation and apply the Lagrangian relaxation technique to scale up the solution approach. Experiments have been performed over the instances generated using the real Singapore transportation network. The results show that we can find significantly better solutions than the deterministic formulation.

Mobility-Driven BLE Transmit-Power Adaptation for Participatory Data Muling

2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS), 2018

This paper analyzes a human-centric framework, called SmartABLE, for easy retrieval of the sensor... more This paper analyzes a human-centric framework, called SmartABLE, for easy retrieval of the sensor values from pervasively deployed smart objects in a campus-like environment. In this framework, smartphones carried by campus occupants act as data mules, opportunistically retrieving data from nearby BLE (Bluetooth Low Energy) equipped smart object sensors and relaying them to a backend repository. We focus specifically on dynamically varying the transmission power of the deployed BLE beacons, so as to extend their operational lifetime without sacrificing the frequency of sensor data retrieval. We propose a memetic algorithm-based power adaptation strategy that can handle deployments of thousands of beacons and tackles two distinct objectives: (1) maximizing BLE beacon lifetime, and (2) reducing the BLE scanning energy of the mules. Using real-world movement traces on the Singapore Management University campus, we show that the benefit of such mule movement-aware power adaptation: it provides reliably frequent retrieval of BLE sensor data, while achieving a significant (5-fold) increase in the sensor lifetime, compared to a traditional fixed-power approach.

Exact and heuristic approaches for the multi-agent orienteering problem with capacity constraints

2017 IEEE Symposium Series on Computational Intelligence (SSCI), 2017

This paper introduces and addresses a new multi-agent variant of the orienteering problem (OP), n... more This paper introduces and addresses a new multi-agent variant of the orienteering problem (OP), namely the multi-agent orienteering problem with capacity constraints (MAOPCC). Different from the existing variants of OP, MAOPCC allows a group of visitors to concurrently visit a node but limits the number of visitors simultaneously being served at each node. In this work, we solve MAOPCC in a centralized manner and optimize the total collected rewards of all agents. A branch and bound algorithm is first proposed to find an optimal MAOPCC solution. Since finding an optimal solution for MAOPCC can become intractable as the number of vertices and agents increases, a computationally efficient sequential algorithm that sacrifices the solution quality is then proposed. Finally, a probabilistic iterated local search algorithm is developed to find a sufficiently good solution in a reasonable time. Our experimental results show that the latter strikes a good tradeoff between the solution quali...

Privacy in context-aware mobile crowdsourcing systems

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Mobile crowd-sourcing can become as a strategy to perform time-sensitive urban tasks (such as mun... more Mobile crowd-sourcing can become as a strategy to perform time-sensitive urban tasks (such as municipal monitoring and last mile logistics) by effectively coordinating smartphone users. The success of the mobile crowd-sourcing platform depends mainly on its effectiveness in engaging crowd-workers, and recent studies have shown that compared to the pull-based approach, which relies on crowd-workers to browse and commit to tasks they would want to perform, the push-based approach can take into consideration of worker's daily routine, and generate highly effective recommendations. As a result, workers waste less time on detours, plan more in advance, and require much less planning effort. However, the push-based systems are not without drawbacks. The major concern is the potential privacy invasion that could result from the disclosure of individual's mobility traces to the crowd-sourcing platform. In this paper, we first demonstrate specific threats of continuous sharing of use...

A Driver Guidance System for Taxis in Singapore

Traditional taxi fleet operators world-over have been facing intense competitions from various ri... more Traditional taxi fleet operators world-over have been facing intense competitions from various ride-hailing services such as Uber and Grab. Based on our studies on the taxi industry in Singapore, we see that the emergence of Uber and Grab in the ride-hailing market has greatly impacted the taxi industry: the average daily taxi ridership for the past two years has been falling continuously, by close to 20% in total. In this work, we discuss how efficient real-time data analytics and large-scale multiagent optimization technology could help taxi drivers compete against more technologically advanced service platforms. Our system has been in field trial with close to 400 drivers, and our initial results show that by following our recommendations, drivers on average save 21.5% on roaming time.

An Experimental Investigation of Product Competition and Marketing in Social Networks

We conduct computational experiment using Facebook data to evaluate competing firms’ initial mark... more We conduct computational experiment using Facebook data to evaluate competing firms’ initial market seeding and subsequent targeted marketing strategies that influence consumers’ new product adoption decisions. We find that firms generally overspend their advertising budget in the market seeding phase. In the subsequent market advertising phase, a coupon strategy (equivalent to price discount) generally yields higher market share than the strategy of distributing free product samples. The effect is more significant when both price and product quality are low. We offer managerial insights into firms’ effective competition strategies for new product introduction in the presence of consumers’ word of mouth effects in social networks.

Fickle Fingers: Ride-Hail Surge Factors and Taxi Bookings

SSRN Electronic Journal, 2018

Exploiting a unique feature of the Singapore taxi and ride-hailing markets, we study whether cons... more Exploiting a unique feature of the Singapore taxi and ride-hailing markets, we study whether consumers switch between taxis and ride-hailing based on their relative prices. We find that a 10% increase in the ride-hailing surge factor raises taxi bookings by 4% in the same origin area in the same half-hour interval. This substitution between ridehailing services to taxi bookings in response to the surge factor comprises around 10% of taxi booking revenues. Further, if a typical customer switches from ride-hailing services to taxi bookings whenever the surge factor is greater than one then she would save around 18% of the overall fare.

Obfuscation At-Source

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2018

By effectively reaching out to and engaging larger population of mobile users, mobile crowd-sourc... more By effectively reaching out to and engaging larger population of mobile users, mobile crowd-sourcing has become a strategy to perform large amount of urban tasks. The recent empirical studies have shown that compared to the pull-based approach, which expects the users to browse through the list of tasks to perform, the push-based approach that actively recommends tasks can greatly improve the overall system performance. As the efficiency of the push-based approach is achieved by incorporating worker's mobility traces, privacy is naturally a concern. In this paper, we propose a novel, 2-stage and user-controlled obfuscation technique that provides a trade off-amenable framework that caters to multi-attribute privacy measures (considering the per-user sensitivity and global uniqueness of locations). We demonstrate the effectiveness of our approach by testing it using the real-world data collected from the well-established TA$Ker platform. More specifically, we show that one can in...

Learning by Driving: Evidence from Taxi Driver Wages in Singapore

SSRN Electronic Journal, 2015

The significance of "Learning by doing" (LBD) in the economy has been disputed by economists. We ... more The significance of "Learning by doing" (LBD) in the economy has been disputed by economists. We use a unique dataset from a Singaporean taxi fleet consisting of 3,250 drivers with over 520 million data points that track cabdrivers' minute-by-minute work routines for two years to test if taxi drivers exhibit LBD. After controlling for individual-level differences, i.e. socio-economics, we find strong evidence of LBD. We document this mechanism and show taxi drivers learn through temporal (when) and locational (where) and technology (booking) optimization. The study has implications for the contribution of LBD behaviour to the economy and its direct impact on individual wages. We also discuss the implication of our findings for taxi markets experiencing a rise in use of technology (i.e. booking systems based on applications), which has recently drawn the interest of policymakers.

Equilibria in Symmetric Games

Upping the Game of Taxi Driving in the Age of Uber

Proceedings of the AAAI Conference on Artificial Intelligence

In most cities, taxis play an important role in providing point-to-point transportation service. ... more In most cities, taxis play an important role in providing point-to-point transportation service. If the taxi service is reliable, responsive, and cost-effective, past studies show that taxi-like services can be a viable choice in replacing a significant amount of private cars. However, making taxi services efficient is extremely challenging, mainly due to the fact that taxi drivers are self-interested and they operate with only local information. Although past research has demonstrated how recommendation systems could potentially help taxi drivers in improving their performance, most of these efforts are not feasible in practice. This is mostly due to the lack of both the comprehensive data coverage and an efficient recommendation engine that can scale to tens of thousands of drivers. In this paper, we propose a comprehensive working platform called the Driver Guidance System (DGS). With real-time citywide taxi data provided by our collaborator in Singapore, we demonstrate how we ca...

Trading agents competing: Performance, progress, and market effectiveness

Since the year 2000, the annual trading agent competition has provided a forum for designers to e... more Since the year 2000, the annual trading agent competition has provided a forum for designers to evaluate programmed trading techniques in a challenging market scenario in competition with other design groups. After three years of apparent progress, we attempt to evaluate the trading competence of competition participants, in the 2002 tournament and over time. Although absolute measure of individual performance is difficult to assess, relative measures, and measures of the market performance overall are more amenable to direct analysis. We quantify the effectiveness of the TAC travel market in terms of allocative efficiency, finding improvement within and between tournaments. By comparison with alternative allocation benchmarks, we can calibrate this efficiency, and identify opportunities for further gain from trade. 1

Density Peaks Clustering Approach for Discovering Demand Hot Spots in City-scale Taxi Fleet Dataset

Abstract—In this paper, we introduce a variant of the density peaks clustering (DPC) approach for... more Abstract—In this paper, we introduce a variant of the density peaks clustering (DPC) approach for discovering demand hot spots from a low-frequency, low-quality taxi fleet operational dataset. From the literature, the DPC approach mainly uses density peaks as features to discover potential cluster centers, and this requires distances between all pairs of data points to be calculated. This implies that the DPC approach can only be applied to cases with relatively small numbers of data points. For the domain of urban taxi operations that we are interested in, we could have millions of demand points per day, and calculating all-pair distances between all demand points would be practically impossible, thus making DPC approach not applicable. To address this issue, we project all points to a density image and execute our variant of the DPC algorithm on the processed image. Experiment results show that our proposed DPC variant could get similar results as original DPC, yet with much short...

TaxiSim: A Multiagent Simulation Platform for Evaluating Taxi Fleet Operations

TaxiSim: A multiagent simulation platform for evaluating taxi fleet operations

A service choice model for optimizing taxi service delivery

Abstract—Taxi service has undergone radical revamp in recent years. In particular, significant in... more Abstract—Taxi service has undergone radical revamp in recent years. In particular, significant investments in commu-nication system and GPS devices have improved quality of taxi services through better dispatches. In this paper, we propose to leverage on such infrastructure and build a service choice model that helps individual drivers in deciding whether to serve a specific taxi stand or not. We demonstrate the value of our model by applying it to a real-world scenario. We also highlight interesting new potential approaches that could significantly improve the quality of taxi services. I.

Achieving Stable and Fair Profit Allocation with Minimum Subsidy in Collaborative Logistics

With the advent of e-commerce, logistics providers are faced with the challenge of handling fluct... more With the advent of e-commerce, logistics providers are faced with the challenge of handling fluctuating and sparsely dis-tributed demand, which raises their operational costs signif-icantly. As a result, horizontal cooperation are gaining mo-mentum around the world. One of the major impediments, however, is the lack of stable and fair profit sharing mecha-nism. In this paper, we address this problem using the frame-work of computational cooperative games. We first present cooperative vehicle routing game as a model for collabora-tive logistics operations. Using the axioms of Shapley value as the conditions for fairness, we show that a stable, fair and budget balanced allocation does not exist in many instances of the game. By relaxing budget balance, we then propose an allocation scheme based on the normalized Shapley value. We show that this scheme maintains stability and fairness while requiring minimum subsidy. Finally, using numerical exper-iments we demonstrate the feasibility ...

TRACCS: Trajectory-aware coordinated urban crowd-sourcing

We investigate the problem of large-scale mobile crowd-tasking, where a large pool of citizen cro... more We investigate the problem of large-scale mobile crowd-tasking, where a large pool of citizen crowd-workers are used to perform a variety of location-specific urban logis-tics tasks. Current approaches to such mobile crowd-tasking are very decentralized: a crowd-tasking platform usually pro-vides each worker a set of available tasks close to the worker’s current location; each worker then independently chooses which tasks she wants to accept and perform. In contrast, we propose TRACCS, a more coordinated task assignment ap-proach, where the crowd-tasking platform assigns a sequence of tasks to each worker, taking into account their expected location trajectory over a wider time horizon, as opposed to just instantaneous location. We formulate such task assign-ment as an optimization problem, that seeks to maximize the total payoff from all assigned tasks, subject to a maximum bound on the detour (from the expected path) that a worker will experience to complete her assigned tasks. We...

Designing Bus Bridging Services for Regular

Abstract. We are concerned with the regular egress problem after a ma-jor event at a known locati... more Abstract. We are concerned with the regular egress problem after a ma-jor event at a known location. Without properly design complementary transport services, such sudden crowd build-ups will overwhelm the ex-isting infrastructure. In this paper, we introduce a novel flow-rate based model to model the dynamic movement of passengers over the trans-portation flow network. Based on this basic model, an integer linear programming model is proposed to solve the bus bridging problem per-manently. We validate our model against a real scenario in Singapore, where a newly constructed mega-stadium hosts various large events reg-ularly. The results show that the proposed approach effectively enables regular egress, and achieves almost 24.1 % travel time reduction with an addition of 40 buses serving 18.7 % of the passengers.

Lagrangian Relaxation for Large-Scale Multi-Agent Planning (Extended Abstract)

Multi-agent planning is a well-studied problem with applications in various areas. Due to computa... more Multi-agent planning is a well-studied problem with applications in various areas. Due to computational constraints, existing research typically focuses either on unstructured domains with many agents, where we are content with heuristic solutions, or domains with small numbers of agents or special structure, where we can find provably near-optimal solutions. In contrast, here we focus on provably near-optimal solutions in domains with many agents, by exploiting influence limits. To that end, we make two key contributions: (a) an algorithm, based on Lagrangian relaxation and randomized rounding, for solving multi-agent planning problems represented as large mixed-integer programs; (b) a proof of convergence of our algorithm to a near-optimal solution.

Multiscale Generative Models: Improving Performance of a Generative Model Using Feedback from Other Dependent Generative Models

Proceedings of the AAAI Conference on Artificial Intelligence

Realistic fine-grained multi-agent simulation of real-world complex systems is crucial for many d... more Realistic fine-grained multi-agent simulation of real-world complex systems is crucial for many downstream tasks such as reinforcement learning. Recent work has used generative models (GANs in particular) for providing high-fidelity simulation of real-world systems. However, such generative models are often monolithic and miss out on modeling the interaction in multi-agent systems. In this work, we take a first step towards building multiple interacting generative models (GANs) that reflects the interaction in real world. We build and analyze a hierarchical set-up where a higher-level GAN is conditioned on the output of multiple lower-level GANs. We present a technique of using feedback from the higher-level GAN to improve performance of lower-level GANs. We mathematically characterize the conditions under which our technique is impactful, including understanding the transfer learning nature of our set-up. We present three distinct experiments on synthetic data, time series data, an...

Multi-Agent Task Assignment for Mobile Crowdsourcing under Trajectory Uncertainties

Adaptive Agents and Multi-Agents Systems, May 4, 2015

In this work, we investigate the problem of mobile crowdsourcing, where workers are financially m... more In this work, we investigate the problem of mobile crowdsourcing, where workers are financially motivated to perform location-based tasks physically. Unlike current industry practice that relies on workers to manually browse and filter tasks to perform, we intend to automatically make task recommendations based on workers' historical trajectories and desired time budgets. However, predicting workers' trajectories is inevitably faced with uncertainties, as no one will take exactly the same route every day; yet such uncertainties are oftentimes abstracted away in the known literature. In this work, we depart from the deterministic modeling and study the stochastic task recommendation problem where each worker is associated with several predicted routine routes with probabilities. We formulate this problem as a stochastic integer linear program whose goal is to maximize the expected total utility achieved by all workers. We further exploit the separable structure of the formulation and apply the Lagrangian relaxation technique to scale up the solution approach. Experiments have been performed over the instances generated using the real Singapore transportation network. The results show that we can find significantly better solutions than the deterministic formulation.

Mobility-Driven BLE Transmit-Power Adaptation for Participatory Data Muling

2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS), 2018

This paper analyzes a human-centric framework, called SmartABLE, for easy retrieval of the sensor... more This paper analyzes a human-centric framework, called SmartABLE, for easy retrieval of the sensor values from pervasively deployed smart objects in a campus-like environment. In this framework, smartphones carried by campus occupants act as data mules, opportunistically retrieving data from nearby BLE (Bluetooth Low Energy) equipped smart object sensors and relaying them to a backend repository. We focus specifically on dynamically varying the transmission power of the deployed BLE beacons, so as to extend their operational lifetime without sacrificing the frequency of sensor data retrieval. We propose a memetic algorithm-based power adaptation strategy that can handle deployments of thousands of beacons and tackles two distinct objectives: (1) maximizing BLE beacon lifetime, and (2) reducing the BLE scanning energy of the mules. Using real-world movement traces on the Singapore Management University campus, we show that the benefit of such mule movement-aware power adaptation: it provides reliably frequent retrieval of BLE sensor data, while achieving a significant (5-fold) increase in the sensor lifetime, compared to a traditional fixed-power approach.

Exact and heuristic approaches for the multi-agent orienteering problem with capacity constraints

2017 IEEE Symposium Series on Computational Intelligence (SSCI), 2017

This paper introduces and addresses a new multi-agent variant of the orienteering problem (OP), n... more This paper introduces and addresses a new multi-agent variant of the orienteering problem (OP), namely the multi-agent orienteering problem with capacity constraints (MAOPCC). Different from the existing variants of OP, MAOPCC allows a group of visitors to concurrently visit a node but limits the number of visitors simultaneously being served at each node. In this work, we solve MAOPCC in a centralized manner and optimize the total collected rewards of all agents. A branch and bound algorithm is first proposed to find an optimal MAOPCC solution. Since finding an optimal solution for MAOPCC can become intractable as the number of vertices and agents increases, a computationally efficient sequential algorithm that sacrifices the solution quality is then proposed. Finally, a probabilistic iterated local search algorithm is developed to find a sufficiently good solution in a reasonable time. Our experimental results show that the latter strikes a good tradeoff between the solution quali...

Privacy in context-aware mobile crowdsourcing systems

2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 2017

Mobile crowd-sourcing can become as a strategy to perform time-sensitive urban tasks (such as mun... more Mobile crowd-sourcing can become as a strategy to perform time-sensitive urban tasks (such as municipal monitoring and last mile logistics) by effectively coordinating smartphone users. The success of the mobile crowd-sourcing platform depends mainly on its effectiveness in engaging crowd-workers, and recent studies have shown that compared to the pull-based approach, which relies on crowd-workers to browse and commit to tasks they would want to perform, the push-based approach can take into consideration of worker's daily routine, and generate highly effective recommendations. As a result, workers waste less time on detours, plan more in advance, and require much less planning effort. However, the push-based systems are not without drawbacks. The major concern is the potential privacy invasion that could result from the disclosure of individual's mobility traces to the crowd-sourcing platform. In this paper, we first demonstrate specific threats of continuous sharing of use...

A Driver Guidance System for Taxis in Singapore

Traditional taxi fleet operators world-over have been facing intense competitions from various ri... more Traditional taxi fleet operators world-over have been facing intense competitions from various ride-hailing services such as Uber and Grab. Based on our studies on the taxi industry in Singapore, we see that the emergence of Uber and Grab in the ride-hailing market has greatly impacted the taxi industry: the average daily taxi ridership for the past two years has been falling continuously, by close to 20% in total. In this work, we discuss how efficient real-time data analytics and large-scale multiagent optimization technology could help taxi drivers compete against more technologically advanced service platforms. Our system has been in field trial with close to 400 drivers, and our initial results show that by following our recommendations, drivers on average save 21.5% on roaming time.

An Experimental Investigation of Product Competition and Marketing in Social Networks

We conduct computational experiment using Facebook data to evaluate competing firms’ initial mark... more We conduct computational experiment using Facebook data to evaluate competing firms’ initial market seeding and subsequent targeted marketing strategies that influence consumers’ new product adoption decisions. We find that firms generally overspend their advertising budget in the market seeding phase. In the subsequent market advertising phase, a coupon strategy (equivalent to price discount) generally yields higher market share than the strategy of distributing free product samples. The effect is more significant when both price and product quality are low. We offer managerial insights into firms’ effective competition strategies for new product introduction in the presence of consumers’ word of mouth effects in social networks.

Fickle Fingers: Ride-Hail Surge Factors and Taxi Bookings

SSRN Electronic Journal, 2018

Exploiting a unique feature of the Singapore taxi and ride-hailing markets, we study whether cons... more Exploiting a unique feature of the Singapore taxi and ride-hailing markets, we study whether consumers switch between taxis and ride-hailing based on their relative prices. We find that a 10% increase in the ride-hailing surge factor raises taxi bookings by 4% in the same origin area in the same half-hour interval. This substitution between ridehailing services to taxi bookings in response to the surge factor comprises around 10% of taxi booking revenues. Further, if a typical customer switches from ride-hailing services to taxi bookings whenever the surge factor is greater than one then she would save around 18% of the overall fare.

Obfuscation At-Source

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2018

By effectively reaching out to and engaging larger population of mobile users, mobile crowd-sourc... more By effectively reaching out to and engaging larger population of mobile users, mobile crowd-sourcing has become a strategy to perform large amount of urban tasks. The recent empirical studies have shown that compared to the pull-based approach, which expects the users to browse through the list of tasks to perform, the push-based approach that actively recommends tasks can greatly improve the overall system performance. As the efficiency of the push-based approach is achieved by incorporating worker's mobility traces, privacy is naturally a concern. In this paper, we propose a novel, 2-stage and user-controlled obfuscation technique that provides a trade off-amenable framework that caters to multi-attribute privacy measures (considering the per-user sensitivity and global uniqueness of locations). We demonstrate the effectiveness of our approach by testing it using the real-world data collected from the well-established TA$Ker platform. More specifically, we show that one can in...

Learning by Driving: Evidence from Taxi Driver Wages in Singapore

SSRN Electronic Journal, 2015

The significance of "Learning by doing" (LBD) in the economy has been disputed by economists. We ... more The significance of "Learning by doing" (LBD) in the economy has been disputed by economists. We use a unique dataset from a Singaporean taxi fleet consisting of 3,250 drivers with over 520 million data points that track cabdrivers' minute-by-minute work routines for two years to test if taxi drivers exhibit LBD. After controlling for individual-level differences, i.e. socio-economics, we find strong evidence of LBD. We document this mechanism and show taxi drivers learn through temporal (when) and locational (where) and technology (booking) optimization. The study has implications for the contribution of LBD behaviour to the economy and its direct impact on individual wages. We also discuss the implication of our findings for taxi markets experiencing a rise in use of technology (i.e. booking systems based on applications), which has recently drawn the interest of policymakers.

Equilibria in Symmetric Games

Upping the Game of Taxi Driving in the Age of Uber

Proceedings of the AAAI Conference on Artificial Intelligence

In most cities, taxis play an important role in providing point-to-point transportation service. ... more In most cities, taxis play an important role in providing point-to-point transportation service. If the taxi service is reliable, responsive, and cost-effective, past studies show that taxi-like services can be a viable choice in replacing a significant amount of private cars. However, making taxi services efficient is extremely challenging, mainly due to the fact that taxi drivers are self-interested and they operate with only local information. Although past research has demonstrated how recommendation systems could potentially help taxi drivers in improving their performance, most of these efforts are not feasible in practice. This is mostly due to the lack of both the comprehensive data coverage and an efficient recommendation engine that can scale to tens of thousands of drivers. In this paper, we propose a comprehensive working platform called the Driver Guidance System (DGS). With real-time citywide taxi data provided by our collaborator in Singapore, we demonstrate how we ca...

Trading agents competing: Performance, progress, and market effectiveness

Since the year 2000, the annual trading agent competition has provided a forum for designers to e... more Since the year 2000, the annual trading agent competition has provided a forum for designers to evaluate programmed trading techniques in a challenging market scenario in competition with other design groups. After three years of apparent progress, we attempt to evaluate the trading competence of competition participants, in the 2002 tournament and over time. Although absolute measure of individual performance is difficult to assess, relative measures, and measures of the market performance overall are more amenable to direct analysis. We quantify the effectiveness of the TAC travel market in terms of allocative efficiency, finding improvement within and between tournaments. By comparison with alternative allocation benchmarks, we can calibrate this efficiency, and identify opportunities for further gain from trade. 1

Density Peaks Clustering Approach for Discovering Demand Hot Spots in City-scale Taxi Fleet Dataset

Abstract—In this paper, we introduce a variant of the density peaks clustering (DPC) approach for... more Abstract—In this paper, we introduce a variant of the density peaks clustering (DPC) approach for discovering demand hot spots from a low-frequency, low-quality taxi fleet operational dataset. From the literature, the DPC approach mainly uses density peaks as features to discover potential cluster centers, and this requires distances between all pairs of data points to be calculated. This implies that the DPC approach can only be applied to cases with relatively small numbers of data points. For the domain of urban taxi operations that we are interested in, we could have millions of demand points per day, and calculating all-pair distances between all demand points would be practically impossible, thus making DPC approach not applicable. To address this issue, we project all points to a density image and execute our variant of the DPC algorithm on the processed image. Experiment results show that our proposed DPC variant could get similar results as original DPC, yet with much short...

TaxiSim: A Multiagent Simulation Platform for Evaluating Taxi Fleet Operations

TaxiSim: A multiagent simulation platform for evaluating taxi fleet operations

A service choice model for optimizing taxi service delivery

Abstract—Taxi service has undergone radical revamp in recent years. In particular, significant in... more Abstract—Taxi service has undergone radical revamp in recent years. In particular, significant investments in commu-nication system and GPS devices have improved quality of taxi services through better dispatches. In this paper, we propose to leverage on such infrastructure and build a service choice model that helps individual drivers in deciding whether to serve a specific taxi stand or not. We demonstrate the value of our model by applying it to a real-world scenario. We also highlight interesting new potential approaches that could significantly improve the quality of taxi services. I.

Achieving Stable and Fair Profit Allocation with Minimum Subsidy in Collaborative Logistics

With the advent of e-commerce, logistics providers are faced with the challenge of handling fluct... more With the advent of e-commerce, logistics providers are faced with the challenge of handling fluctuating and sparsely dis-tributed demand, which raises their operational costs signif-icantly. As a result, horizontal cooperation are gaining mo-mentum around the world. One of the major impediments, however, is the lack of stable and fair profit sharing mecha-nism. In this paper, we address this problem using the frame-work of computational cooperative games. We first present cooperative vehicle routing game as a model for collabora-tive logistics operations. Using the axioms of Shapley value as the conditions for fairness, we show that a stable, fair and budget balanced allocation does not exist in many instances of the game. By relaxing budget balance, we then propose an allocation scheme based on the normalized Shapley value. We show that this scheme maintains stability and fairness while requiring minimum subsidy. Finally, using numerical exper-iments we demonstrate the feasibility ...

TRACCS: Trajectory-aware coordinated urban crowd-sourcing

We investigate the problem of large-scale mobile crowd-tasking, where a large pool of citizen cro... more We investigate the problem of large-scale mobile crowd-tasking, where a large pool of citizen crowd-workers are used to perform a variety of location-specific urban logis-tics tasks. Current approaches to such mobile crowd-tasking are very decentralized: a crowd-tasking platform usually pro-vides each worker a set of available tasks close to the worker’s current location; each worker then independently chooses which tasks she wants to accept and perform. In contrast, we propose TRACCS, a more coordinated task assignment ap-proach, where the crowd-tasking platform assigns a sequence of tasks to each worker, taking into account their expected location trajectory over a wider time horizon, as opposed to just instantaneous location. We formulate such task assign-ment as an optimization problem, that seeks to maximize the total payoff from all assigned tasks, subject to a maximum bound on the detour (from the expected path) that a worker will experience to complete her assigned tasks. We...

Designing Bus Bridging Services for Regular

Abstract. We are concerned with the regular egress problem after a ma-jor event at a known locati... more Abstract. We are concerned with the regular egress problem after a ma-jor event at a known location. Without properly design complementary transport services, such sudden crowd build-ups will overwhelm the ex-isting infrastructure. In this paper, we introduce a novel flow-rate based model to model the dynamic movement of passengers over the trans-portation flow network. Based on this basic model, an integer linear programming model is proposed to solve the bus bridging problem per-manently. We validate our model against a real scenario in Singapore, where a newly constructed mega-stadium hosts various large events reg-ularly. The results show that the proposed approach effectively enables regular egress, and achieves almost 24.1 % travel time reduction with an addition of 40 buses serving 18.7 % of the passengers.

Lagrangian Relaxation for Large-Scale Multi-Agent Planning (Extended Abstract)

Multi-agent planning is a well-studied problem with applications in various areas. Due to computa... more Multi-agent planning is a well-studied problem with applications in various areas. Due to computational constraints, existing research typically focuses either on unstructured domains with many agents, where we are content with heuristic solutions, or domains with small numbers of agents or special structure, where we can find provably near-optimal solutions. In contrast, here we focus on provably near-optimal solutions in domains with many agents, by exploiting influence limits. To that end, we make two key contributions: (a) an algorithm, based on Lagrangian relaxation and randomized rounding, for solving multi-agent planning problems represented as large mixed-integer programs; (b) a proof of convergence of our algorithm to a near-optimal solution.