Scheduling with Predictions and the Price of Misprediction

TL;DR

This paper analyzes how predictions of job service times, often generated by machine learning, impact scheduling performance and introduces the concept of the 'price of misprediction' to quantify the cost of prediction errors.

Contribution

It derives performance formulae for scheduling strategies that utilize predicted service times and introduces the 'price of misprediction' framework to measure prediction costs.

Findings

Derived formulae for scheduling performance with predictions

Introduced the 'price of misprediction' concept

Quantified the impact of prediction errors on scheduling efficiency

Abstract

In many traditional job scheduling settings, it is assumed that one knows the time it will take for a job to complete service. In such cases, strategies such as shortest job first can be used to improve performance in terms of measures such as the average time a job waits in the system. We consider the setting where the service time is not known, but is predicted by for example a machine learning algorithm. Our main result is the derivation, under natural assumptions, of formulae for the performance of several strategies for queueing systems that use predictions for service times in order to schedule jobs. As part of our analysis, we suggest the framework of the "price of misprediction," which offers a measure of the cost of using predicted information.