Lazy clause generation in retrospect

Olga Ohrimenko; Peter J. Stuckey; Michael Codish

doi:10.1007/s10601-026-09390-9

Lazy clause generation in retrospect

Olga Ohrimenko
, Peter J. Stuckey
, Michael Codish

Institute for the Theory of Computing

Research output: Contribution to journal › Article › peer-review

Abstract

Lazy clause generation (LCG) is an approach to building constraint programming (CP) solvers that uses ideas from Boolean Satisfiability (SAT) solvers. Propagation in the CP solver is extended to explain itself, lazily creating explanation clauses. This allows SAT style conflict analysis to be used to create clausal nogoods that record the reason for failure, and avoid the same failure in the future. LCG has proven to be a highly effective architecture, used in all most performant CP solvers. In this review we consider how LCG has been refined and expanded over the years since its invention.

Original language	English
Journal	Constraints
DOIs	https://doi.org/10.1007/s10601-026-09390-9
State	Accepted/In press - 1 Jan 2026

Keywords

Discrete optimisation
Nogood learning
Search

ASJC Scopus subject areas

Software
Discrete Mathematics and Combinatorics
Computational Theory and Mathematics
Artificial Intelligence

Access to Document

10.1007/s10601-026-09390-9

Cite this

@article{3b1f583c38ab409ebb7456f2a089bc5b,

title = "Lazy clause generation in retrospect",

abstract = "Lazy clause generation (LCG) is an approach to building constraint programming (CP) solvers that uses ideas from Boolean Satisfiability (SAT) solvers. Propagation in the CP solver is extended to explain itself, lazily creating explanation clauses. This allows SAT style conflict analysis to be used to create clausal nogoods that record the reason for failure, and avoid the same failure in the future. LCG has proven to be a highly effective architecture, used in all most performant CP solvers. In this review we consider how LCG has been refined and expanded over the years since its invention.",

keywords = "Discrete optimisation, Nogood learning, Search",

author = "Olga Ohrimenko and Stuckey, \{Peter J.\} and Michael Codish",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2026.",

year = "2026",

month = jan,

day = "1",

doi = "10.1007/s10601-026-09390-9",

language = "English",

journal = "Constraints",

issn = "1383-7133",

publisher = "Springer",

}

TY - JOUR

T1 - Lazy clause generation in retrospect

AU - Ohrimenko, Olga

AU - Stuckey, Peter J.

AU - Codish, Michael

PY - 2026/1/1

Y1 - 2026/1/1

N2 - Lazy clause generation (LCG) is an approach to building constraint programming (CP) solvers that uses ideas from Boolean Satisfiability (SAT) solvers. Propagation in the CP solver is extended to explain itself, lazily creating explanation clauses. This allows SAT style conflict analysis to be used to create clausal nogoods that record the reason for failure, and avoid the same failure in the future. LCG has proven to be a highly effective architecture, used in all most performant CP solvers. In this review we consider how LCG has been refined and expanded over the years since its invention.

AB - Lazy clause generation (LCG) is an approach to building constraint programming (CP) solvers that uses ideas from Boolean Satisfiability (SAT) solvers. Propagation in the CP solver is extended to explain itself, lazily creating explanation clauses. This allows SAT style conflict analysis to be used to create clausal nogoods that record the reason for failure, and avoid the same failure in the future. LCG has proven to be a highly effective architecture, used in all most performant CP solvers. In this review we consider how LCG has been refined and expanded over the years since its invention.

KW - Discrete optimisation

KW - Nogood learning

KW - Search

UR - https://www.scopus.com/pages/publications/105044362761

U2 - 10.1007/s10601-026-09390-9

DO - 10.1007/s10601-026-09390-9

M3 - Article

AN - SCOPUS:105044362761

SN - 1383-7133

JO - Constraints

JF - Constraints

ER -

Lazy clause generation in retrospect

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this