Safe Approximations of Chance Constraints Using Historical Data

I. Yanikoglu, D. den Hertog

Research output: Working paperDiscussion paperOther research output

264 Downloads (Pure)

Abstract

This paper proposes a new way to construct uncertainty sets for robust optimization. Our approach uses the available historical data for the uncertain parameters and is based on goodness-of-fit statistics. It guarantees that the probability that the uncertain constraint holds is at least the prescribed value. Compared to existing safe approximation methods for chance constraints, our approach directly uses the historical-data information and leads to tighter uncertainty sets and therefore to better objective values. This improvement is significant especially when the number of uncertain parameters is low. Other advantages of our approach are that it can handle joint chance constraints easily, it can deal with uncertain parameters that are dependent, and it can be extended to nonlinear inequalities. Several numerical examples illustrate the validity of our approach.
Original languageEnglish
Place of PublicationTilburg
PublisherOperations research
Number of pages39
Volume2011-137
Publication statusPublished - 2011

Publication series

NameCentER Discussion Paper
Volume2011-137

Fingerprint

Statistics
Uncertainty

Keywords

  • robust optimization
  • chance constraint
  • phi-divergence
  • goodness-of-fit statistics

Cite this

Yanikoglu, I., & den Hertog, D. (2011). Safe Approximations of Chance Constraints Using Historical Data. (CentER Discussion Paper; Vol. 2011-137). Tilburg: Operations research.
Yanikoglu, I. ; den Hertog, D. / Safe Approximations of Chance Constraints Using Historical Data. Tilburg : Operations research, 2011. (CentER Discussion Paper).
@techreport{ab77f6f2248a42f1bde10050015aec07,
title = "Safe Approximations of Chance Constraints Using Historical Data",
abstract = "This paper proposes a new way to construct uncertainty sets for robust optimization. Our approach uses the available historical data for the uncertain parameters and is based on goodness-of-fit statistics. It guarantees that the probability that the uncertain constraint holds is at least the prescribed value. Compared to existing safe approximation methods for chance constraints, our approach directly uses the historical-data information and leads to tighter uncertainty sets and therefore to better objective values. This improvement is significant especially when the number of uncertain parameters is low. Other advantages of our approach are that it can handle joint chance constraints easily, it can deal with uncertain parameters that are dependent, and it can be extended to nonlinear inequalities. Several numerical examples illustrate the validity of our approach.",
keywords = "robust optimization, chance constraint, phi-divergence, goodness-of-fit statistics",
author = "I. Yanikoglu and {den Hertog}, D.",
note = "Pagination: 39",
year = "2011",
language = "English",
volume = "2011-137",
series = "CentER Discussion Paper",
publisher = "Operations research",
type = "WorkingPaper",
institution = "Operations research",

}

Yanikoglu, I & den Hertog, D 2011 'Safe Approximations of Chance Constraints Using Historical Data' CentER Discussion Paper, vol. 2011-137, Operations research, Tilburg.

Safe Approximations of Chance Constraints Using Historical Data. / Yanikoglu, I.; den Hertog, D.

Tilburg : Operations research, 2011. (CentER Discussion Paper; Vol. 2011-137).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Safe Approximations of Chance Constraints Using Historical Data

AU - Yanikoglu, I.

AU - den Hertog, D.

N1 - Pagination: 39

PY - 2011

Y1 - 2011

N2 - This paper proposes a new way to construct uncertainty sets for robust optimization. Our approach uses the available historical data for the uncertain parameters and is based on goodness-of-fit statistics. It guarantees that the probability that the uncertain constraint holds is at least the prescribed value. Compared to existing safe approximation methods for chance constraints, our approach directly uses the historical-data information and leads to tighter uncertainty sets and therefore to better objective values. This improvement is significant especially when the number of uncertain parameters is low. Other advantages of our approach are that it can handle joint chance constraints easily, it can deal with uncertain parameters that are dependent, and it can be extended to nonlinear inequalities. Several numerical examples illustrate the validity of our approach.

AB - This paper proposes a new way to construct uncertainty sets for robust optimization. Our approach uses the available historical data for the uncertain parameters and is based on goodness-of-fit statistics. It guarantees that the probability that the uncertain constraint holds is at least the prescribed value. Compared to existing safe approximation methods for chance constraints, our approach directly uses the historical-data information and leads to tighter uncertainty sets and therefore to better objective values. This improvement is significant especially when the number of uncertain parameters is low. Other advantages of our approach are that it can handle joint chance constraints easily, it can deal with uncertain parameters that are dependent, and it can be extended to nonlinear inequalities. Several numerical examples illustrate the validity of our approach.

KW - robust optimization

KW - chance constraint

KW - phi-divergence

KW - goodness-of-fit statistics

M3 - Discussion paper

VL - 2011-137

T3 - CentER Discussion Paper

BT - Safe Approximations of Chance Constraints Using Historical Data

PB - Operations research

CY - Tilburg

ER -

Yanikoglu I, den Hertog D. Safe Approximations of Chance Constraints Using Historical Data. Tilburg: Operations research. 2011. (CentER Discussion Paper).