Grant: Engage Grants Program

In today's intensive competitive environment in the oil and gas industry, it is important to increase efficiency inproduction through resource scheduling. Currently, resource scheduling is done manually by rule of thumb,with spreadsheets or commercial tools. These existing approaches for resource scheduling cannot adaptivelyreact to emergent conditions or disturbances such as safety orders from external factors, and cannot efficientlyschedule daily schedulable tasks. As a result, delays in work occur causing service providers to charge standbyfees for people associated with the service to wait for work to start again. Thus, an innovative approach isneeded to help onsite decision makers efficiently plan and schedule the daily operational tasks around well-sitejobs, to minimize the overall costs. However, there are challenges that impede this process:(1) the dynamic and large-scale nature of resource scheduling where there are thousands of wells makes thescheduling process and the prediction of resource requests slow, tedious, and difficult;(2) ineffective data management prevents the acquisition and aggregation of the data about thecontext and operational features of well planning and rigging planning.This research project attempts to address these issues through identifying and using techniques from two majorresearch domains, namely Multiagent System (MAS) technologies and Big Data Analysis. The coordinationmechanisms such as market-inspired coordination mechanism and interaction protocols such as Contract NetProtocol (CNP) in MAS technologies will enable flexible and efficient resource allocation that will lead to anintelligent resource scheduling system. Big Data Analysis will provide the ability to dynamically extract andembed data-driven knowledge to the agents operating in the well scheduling environment to perform predictionand next-move decision-making.