Task Assignment and Scheduling for Some Location-based Services

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Task Assignment and Scheduling for Some Location-based Services"

By

Mr. Peng CHENG


Abstract

Recently, Location-based services (LBSs) refer to the services that are based 
on the location (spatial) data, which bring conveniences to our daily life and 
challenges to both industry and academia. LBSs include "ridesharing" service, 
which arranges rides to vehicles with empty seats based on the locations of the 
riders and the vehicles, the "spatial crowdsourcing" service,  which allows 
requesters to post spatial tasks to specified locations then the crowd workers 
will move to the locations of the assigned tasks to conduct, the 
"location-based mobile advertising" service, which helps the vendors to push 
advertisements to potential customers near to the shops, and the "search 
nearby" service, which queries some points of interests (POI, e.g., hotels, 
parks, museums) near a location. Among the LBSs, online-to-offline (O2O) is a 
widely applied mechanism, where users join activities, plan travel routes and 
order goods online, then perform the according actions offline. To support this 
fundamental mechanism, task assignment and scheduling are necessary and 
important, which match or schedule the tasks to users under the constraints 
(e.g., spatial-temporal constraints, capacity constraints, budget constraints).

In this thesis, we study task assignment and scheduling techniques in three 
practical problems in the ridesharing area and the spatial crowdsourcing area, 
namely the Utility-Aware Ridesharing on Road Networks problem, which matches 
riders to vehicles and schedules the routes for vehicles with a goal of 
maximizing the overall utility of riders (i.e., vehicle-related utility, 
rider-related utility and trajectory-related utility) subject to the 
constraints of the deadlines of riders and the capacities of the vehicles, the 
Reliable Diversity-Based Spatial Crowdsourcing problem, which assigns spatial 
workers to spatial tasks to maximize the completion reliability and the 
spatial/temporal diversities of spatial tasks subject to the constraints of the 
valid periods of tasks and the working areas of the workers, and the 
Multi-Skill Spatial Crowdsourcing problem, which assigns spatial workers to 
multi-skill required tasks to finish as many tasks as possible and to minimize 
the total travel cost of workers.


Date:			Friday, 18 August 2017

Time:			3:00pm - 5:00pm

Venue:			Room 2611
 			Lifts 31/32

Chairman:		Prof. Ying-Ju Chen (ISOM)

Committee Members:	Prof. Lei Chen (Supervisor)
 			Prof. Qiong Luo
 			Prof. Ke Yi
 			Prof. Jiheng Zhang (IELM)
 			Prof. Shuai Ma (Comp., Sci. & Engg., Beihang Univ)


**** ALL are Welcome ****