Human-computer Cooperative Task Scheduling Optimization In Bulk Commodity Exchange Market Supervision And Service Pattern

In recent years,e-commerce trading of bulk commodities has been strongly supported by the state,promoting the development of Chinese economic market.However,at the present stage,regulatory resources are insufficient,and regulatory efficiency is difficult to match the speed of market development.Some unscrupulous elements take advantage of regulatory loopholes and cause a large number of abnormal trading events in a short period of time,disturbing the market order in pursuit of their own interests.In order to effectively supervise the frequent abnormal trading events,regulatory authorities urgently need an efficient regulatory pattern that can make use of limited regulatory resources and enhance regulatory efficiency to achieve the purpose of safeguarding the healthy and stable development of the bulk commodity trading market.At this stage,the regulatory pattern adopted by the regulator is to intervene and supervise after the occurrence of abnormal events.Due to the lack of effective early warning and supervision in the pre-and mid-term of trading,the human and computing resources of trading platforms and the regulator are not effectively utilized,resulting in inefficient human-computer collaboration and wasted resources.Currently,the development of intelligent regulation and service patterns,models and mechanisms can provide early warning and response to abnormal trading events in advance,which poses two new challenges to the human-computer collaboration pattern of regulation and service: 1)how to design execution strategies for regulators to improve the efficiency of regulation;2)how to allocate limited human-computer resources to numerous regulatory links to improve the efficiency of human-computer collaboration.Therefore,this thesis addresses these two difficulties,conducts theoretical research and experimental validation on the execution strategy of regulatory links and the scheduling allocation method of regulatory resources respectively,proposes a human-computer collaborative task scheduling optimization technique for the regulation and service pattern of bulk commodity trading market,and builds a human-computer collaborative task scheduling method and system for functional validation.Firstly,this thesis constructs a human-computer collaborative execution strategy of warehouse receipt supervision for the specific supervisory link of warehouse receipt field inspection,and plans the field inspection paths to different warehouses for the supervisors of trading platform,so that they can conduct as many field inspections of the same warehouse as possible with banks or market regulators at the same time.In this way,the joint field inspection of warehouse receipts supervision by multiple parties is constructed to reduce the possibility of false opening of warehouse receipts,duplicate pledge of warehouse receipts and other problems bringing huge risks to bulk commodity trading.This thesis systematically analyzes three situations in real scenarios: 1)for the small-scale scenario with single supervisor,an exact algorithm based on branch pricing is proposed;2)for the larger-scale scenario with single supervisor,a heuristic method for fast path construction is proposed;3)for the multiple supervisors scenario,a heuristic method based on group clustering and swap reorganization is proposed for task assignment.The specific path construction is carried out after the assignment is completed.The algorithm proposed in this thesis and several multi-agent planning algorithms are compared and tested under several real and simulated data sets,and the results show that the algorithm proposed in this thesis has better optimization effects in terms of total joint checking times and joint checking rates.Then,for the characteristics of many regulatory links and limited resources in the bulk commodity trading market,this thesis proposes a collaborative human-computer scheduling and allocation method for regulatory resources,which centralizes and allocates the regulatory resources of trading platforms and regulatory agencies to different regulatory links,so that as many tasks as possible can obtain resources in time and be completed before the deadline,reducing the risk and loss caused by untimely regulation.In this thesis,we propose a scheduling allocation algorithm based on deadline division and cumulative delay risk weighting for the scenario of periodic tasks;for the scenario of real-time online arrival of tasks,we propose an online scheduling framework to allocate regulatory resources in real time through algorithm modules such as task deadline division,resource allocation and feedback adjustment of execution results,so as to reduce the delay caused by the uncertain completion time of the task.A large number of comparison experiments show that the algorithm proposed in this thesis has better optimization effects in terms of task completion rate,resource utilization and average task waiting time compared with other algorithms.Finally,based on the theoretical research,a prototype system of human-computer collaborative scheduling for supervising resources in bulk commodity trading market is designed and implemented in this thesis.The system adopts an online scheduling framework based on task execution result feedback regulation,implements a friendly interactive interface and complete functional modules,including: regulatory task management,regulatory resource management,scheduling allocation and task execution result feedback modules.In this thesis,the system is tested under three scenarios: task execution result feedback,new operator and idle operator.The test results show that the system is fully functional and stable,and can generate a scheduling allocation scheme with low delay risk and high robustness.