Research On Multitasking Scheduling Optimization Under Different Job Attributes

With the change of domestic and international markets,the domestic manufacturing industry is weak in technology development and innovation,and the enterprises are lack of vitality.The transformation of the manufacturing industry has become urgent.Lean production,flexible manufacturing,intelligent manufacturing and other production and manufacturing modes have also emerged as the times require.The concept of "multi skilled worker" advocated in the cellular production mode proposes that every worker must master as many processes and skills as possible,which coincides with the current multitasking processing mode.With the acceleration of the information process,the phenomenon of multitasking is not uncommon,so a multitasking scheduling model was born.Combined with the current background of sharing economy,considering the factors of resource allocation,different job levels and outsourcing,a new problem model is proposed,which is more practical and has not only important academic value but also strong practical significance.The research contents and innovations of this paper are as follows:Firstly,the resource allocatable multitasking scheduling problem with the same level is studied.Two objectives about time window correlation function are discussed.Linear resource function and convex resource function are considered respectively.A polynomial time algorithm with complexity O(n3)is given.Then,the deterioration effect is considered,and the polynomial time algorithm is also given for the goal of common due date.In particular,when the deterioration coefficients of the job are the same,a more efficient polynomial time algorithm with complexity O O(nlogn)is given.The proposed model realizes the coordination of production efficiency and resource consumption cost.Secondly,the multitasking scheduling problem with different levels is studied.First,considering that the job has two different levels,the high level job is in multitasking environment,and the low level job is in non multitasking environment,this paper studies a multitasking scheduling model in which part of the jobs can not be disturbed,and considers four objectives related to minimizing the maximum completion time,minimizing the total completion time,minimizing the weighted completion time,and minimizing the common due date.The optimal algorithm is given for the first three objectives,for the last goal,a heuristic algorithm is designed,and the upper bound of the ratio to the optimal solution is given.Then,considering that high-level j obs can disturb low-level jobs,while lowlevel jobs cannot disturb high-level jobs,high-level machines can process highlevel and low-level jobs,and low-level machines can only process low-level jobs,this paper studies two multitasking scheduling models with service levels,with the goal of minimizing the maximum completion time,designs its online algorithm,and gives its online competition ratio.The proposed model provides decision support for enterprises to make reasonable production plans and maximizes customer satisfaction.Finally,this paper studies the multitasking scheduling problem with outsourcing in the same level.Considering the background of sharing economy,it is considered that all jobs can be processed by manufacturers themselves,or paid a certain outsourcing cost to be processed by other manufacturers through outsourcing.Firstly,the multitasking scheduling model of single agent outsourcing is considered,and the three objectives of minimizing the maximum completion time,minimizing the total completion time and minimizing the weighted completion time are studied.The pseudo polynomial dynamic programming algorithm is designed respectively.Then,it is assumed that the task is completed by two agents together,and the multitasking scheduling model of double agent outsourcing is studied.For the two cases of non interruptible and interruptible agents,the two objectives of minimizing the maximum completion time and minimizing the total completion time are considered respectively,and the pseudo polynomial dynamic programming algorithm is also given.This model provides a new idea and method for the optimal allocation of enterprise processing resources.The research considers the scheduling optimization problem in which the jobs are hierarchical and factors such as resource allocation,deterioration effect and outsourcing are taken into account,which is not only conducive to reducing the negative impact of multi task behavior and reducing the operating costs of enterprises,but also conducive to strengthening the cooperation between enterprises and promoting the redistribution of market resources.