Energy efficient scheduling of industrial process by reducing idle time and tardiness of jobs

A computer-implemented method of scheduling jobs in an industrial process, the method comprising:

• scheduling, by a computing system, a plurality of jobs using a job scheduling process continually executed at predetermined intervals, the job scheduling process comprising:
• identifying one or more jobs to be executed on a plurality available machines within a manufacturing facility;
• generating a job schedule based on an optimization function that minimizes total energy cost for all the available machines during a time horizon based on a summation of energy cost at each time step between a start time and an end time, wherein the energy cost at each time step is a summation of
• (a) a first energy cost associated with each available machine in sleeping mode during the time step,
• (b) a second energy cost associated with each available machine in stand-by mode during the time step, and
• (c) a third energy cost associated with each available machine in processing mode during the time step;
• wherein one or more scheduling constraints are applied to the optimization function to maximize energy efficiency
• wherein the optimization function is parallelized to perform operations in parallel, comprising:
• mapping a computation or a series of computations of the optimization function onto a grid structure, each grid structure containing multiple thread blocks and each thread block containing multiple threads, wherein thread blocks are selected and configured to optimize job scheduling, wherein each thread has one or more levels of memory access including at least one of:
• a register assigned to a particular thread, a shared memory shared by threads of a thread block, and a device memory shared by all thread blocks, wherein the optimization function is coded to optimize utilization of memory access to primarily utilize the register and the shared memory and to limit device memory utilization to move data in and out of a thread block;
• and
• executing the plurality of jobs on the plurality of available machines based on the job schedule.

A system for scheduling jobs for an industrial process, the system comprising:

• a computer-readable storage medium storing a plurality of jobs to be executed on a plurality machines within a manufacturing facility;
• one or more processors configured to:
• generate a job schedule based on an optimization function that minimizes total energy cost for all the machines during a time horizon based on a summation of energy cost at each time step between a start time and an end time, wherein the energy cost at each time step is a summation of:
• (a) a first energy cost associated with each machine in sleeping mode during the time step,
• (b) a second energy cost associated with each machine in stand-by mode during the time step, and
• (c) a third energy cost associated with each machine in processing mode during the time step;
• wherein one or more scheduling constraints are applied to the optimization function to maximize energy efficiency;
• wherein the optimization function is parallelized to perform operations in parallel, comprising:
• mapping a computation or a series of computations of the optimization function onto a grid structure, each grid structure containing multiple thread blocks and each thread block containing multiple threads, wherein thread blocks are selected and configured to optimize job scheduling, wherein each thread has one or more levels of memory access including at least one of:
• a register assigned to a particular thread, a shared memory shared by threads of a thread block, and a device memory shared by all thread blocks, wherein the optimization function is coded to optimize utilization of memory access to primarily utilize the register and the shared memory and to limit device memory utilization to move data in and out of a thread block;
• and execute the plurality of jobs on the plurality of machines based on the job schedule.