Iterative refinement

Iterative refinement is a method used to improve the accuracy of a model or algorithm by repeatedly refining the sample of observations used to fit the model. The procedure involves progressively adjusting the model based on new data until a satisfactory level of accuracy is achieved.From: Applied Mathematical Methods for Chemical Engineers [2019], A data-driven iterative refinement approach for estimating clearing functions from simulation models of production systems [2019]

Iterative Methods

View Chapter

Purchase Book

Published in Jeffery J. Leader, Numerical Analysis and Scientific Computation, 2022

Jeffery J. Leader

and in principle we can solve Eq. (3.1) for e and set x=x0+e to determine x. In practice of course we will have error in e as well but even still x1=x0+e may be an improved estimate of x. This technique is called iterative refinement(or iterative improvement). It uses the current x0 to predict a correction e to be applied to it.

A data-driven iterative refinement approach for estimating clearing functions from simulation models of production systems

View Article

Journal Information

Published in International Journal of Production Research, 2019

Karthick Gopalswamy, Reha Uzsoy

The intuition behind the Iterative Refinement procedure is to progressively refine the sample of observations used to fit the CF. As the parameters of the CF used in the planning model approach those of the correct CF, the distribution of the observations generated in Step 4 should approach the distribution of the observations under the correct CF. The procedure starts by assuming an uncapacitated planning model that sets releases equal to demand. Constraint (14) ensures that expected output does not exceed the expected capacity of the system, while (15) ensures that the output in any period cannot exceed the planned workload. The original CF (1) of Karmarkar (1989) includes both these constraints, but we do not include them in the LS fitting problem due to the need for constrained regression to enforce these, instead adding them in the planning model where they do not materially increase the complexity of the linear programme.

Iterative refinement

Explore chapters and articles related to this topic

Iterative Methods

A data-driven iterative refinement approach for estimating clearing functions from simulation models of production systems