### Refine

#### Year of publication

#### Document Type

- ZIB-Report (38)
- In Proceedings (16)
- Article (8)
- In Collection (2)
- Book chapter (1)
- Master's Thesis (1)

#### Keywords

- constraint integer programming (10)
- cumulative constraint (6)
- mixed integer programming (6)
- constraint programming (5)
- MIP (4)
- Mixed Integer Programming (4)
- branch-and-cut (4)
- integer programming (4)
- Online Optimization (3)
- Pseudo-Boolean (3)

#### Institute

Constraint Integer Programming (CIP) is a generalization of mixed-integer programming (MIP) in the direction of constraint programming (CP) allowing the inference techniques that have traditionally been the core of \P to be integrated with the problem solving techniques that form the core of complete MIP solvers. In this paper, we investigate the application of CIP to scheduling problems that require resource and start-time assignments to satisfy resource capacities. The best current approach to such problems is logic-based Benders decomposition, a manual decomposition method. We present a CIP model and demonstrate that it achieves performance competitive to the decomposition while out-performing the standard MIP and CP formulations.

Despite the success of constraint programming (CP) for scheduling, the much wider penetration of mixed integer programming (MIP) technology into business applications means that many practical scheduling problems are being addressed with MIP, at least as an initial approach. Furthermore, there has been impressive and well-documented improvements in the power of generic MIP solvers over the past decade.
We empirically demonstrate that on an existing set of resource allocation and scheduling problems standard MIP and CP models are now competitive with the state-of-the-art manual decomposition approach. Motivated by this result, we formulate two tightly coupled hybrid models based on constraint integer programming (CIP) and demonstrate that these models,
which embody advances in CP and MIP, are able to out-perform the CP, MIP, and decomposition models. We conclude that both MIP and CIP
are technologies that should be considered along with CP for solving scheduling problems.

この論文ではソフトウェア・パッケージSCIP Optimization Suite を紹介し，その３つの構成要素：モデリン
グ言語Zimpl, 線形計画（LP: linear programming) ソルバSoPlex, そして，制約整数計画(CIP: constraint
integer programming) に対するソフトウェア・フレームワークSCIP, について述べる．本論文では，この３つの
構成要素を利用して，どのようにして挑戦的な混合整数線形計画問題(MIP: mixed integer linear optimization
problems) や混合整数非線形計画問題(MINLP: mixed integer nonlinear optimization problems) をモデル化
し解くのかを説明する．SCIP は，現在，最も高速なMIP,MINLP ソルバの１つである．いくつかの例により，
Zimpl, SCIP, SoPlex の利用方法を示すとともに，利用可能なインタフェースの概要を示す．最後に，将来の開
発計画の概要について述べる．

This paper introduces the SCIP Optimization Suite and discusses the capabilities of its three components: the modeling language Zimpl, the linear programming solver SoPlex, and the constraint integer programming framework SCIP. We explain how these can be used in concert to model and solve challenging mixed integer linear and nonlinear optimization problems. SCIP is currently one of the fastest non-commercial MIP and MINLP solvers. We demonstrate the usage of Zimpl, SCIP, and SoPlex by selected examples, we give an overview of available interfaces, and outline plans for future development.

Recent improvements using constraint integer programming for resource allocation and scheduling
(2013)

Recently, we compared the performance of mixed-integer programming (MIP), constraint programming (CP), and constraint integer programming (CIP) to a state-of-the-art logic-based Benders manual decomposition (LBBD) for a resource allocation/scheduling problem. For a simple linear relaxation, the LBBD and CIP models deliver comparable performance with MIP also performing well.
Here we show that algorithmic developments in CIP plus the use of an existing tighter relaxation substantially improve one of the CIP approaches. Furthermore, the use of the same relaxation in LBBD and MIP models significantly improves their performance. While such a result is known for LBBD, to the best of our knowledge, the other results are novel. Our experiments show that both CIP and MIP approaches are competitive with LBBD in terms of the number of problems solved to proven optimality, though MIP is about three times slower on average. Further, unlike the LBBD and CIP approaches, the MIP model is able to obtain provably high-quality solutions for all problem instances.

We provide a computational study of the performance of a state-of-the-art solver for nonconvex mixed-integer quadratically constrained programs (MIQCPs). Since successful general-purpose solvers for large problem classes necessarily comprise a variety of algorithmic techniques, we focus especially on the impact of the individual solver components. The solver SCIP used for the experiments implements a branch-and-cut algorithm based on a linear relaxation to solve MIQCPs to global optimality. Our analysis is based on a set of 86 publicly available test instances.

制約整数計画ソルバ SCIP の並列化
(2013)

制約整数計画(CIP: Constraint Integer Programming)は，制約プログラミング(CP: Constraint Programming)，
混合整数計画(MIP: Mixed Integer Programming), 充足可能性問題(SAT: Satisfiability Problems)の研究
分野におけるモデリング技術と解法を統合している．その結果，制約整数計画は，広いクラスの最適化問題を
扱うことができる．SCIP (Solving Constraint Integer Programs)は，CIPを解くソルバとして実装され,
Zuse Institute Berlin (ZIB)の研究者を中心として継続的に拡張が続けられている．本論文では，
著者らによって開発されたSCIP に対する2種類の並列化拡張を紹介する．
一つは，複数計算ノード間で大規模に並列動作するParaSCIP である．
もう一つは，複数コアと共有メモリを持つ１台の計算機上で(スレッド)並列で動作するFiberSCIP である．
ParaSCIP は，HLRN IIスーパーコンピュータ上で，
一つのインスタンスを解くために最大7,168 コアを利用した動作実績がある．また，
統計数理研究所のFujitsu PRIMERGY RX200S5上でも，最大512コアを利用した動作実績がある．
統計数理研究所のFujitsu PRIMERGY RX200S5上 では，これまでに最適解が得られていなかった
MIPLIB2010のインスタンスであるdg012142に最適解を与えた．

Dual presolving reductions are a class of reformulation techniques that remove feasible or even optimal solutions while guaranteeing that at least one optimal solution remains, as long as the original problem was feasible. Presolving and dual reductions are important components of state-of-the-art mixed-integer linear programming solvers. In this paper, we introduce them both as unified, practical concepts in constraint programming solvers. Building on the existing idea of variable locks, we formally define and justify the use of dual information for cumulative constraints during a presolving phase of a solver. In particular, variable locks are used to decompose cumulative constraints, detect irrelevant variables, and infer variable assignments and domain reductions. Since the computational complexity of propagation algorithms typically depends on the number of variables and/or domain size, such dual reductions are a source of potential computational speed-up. Through experimental evidence on resource constrained project scheduling problems, we demonstrate that the conditions for dual reductions are present in well-known benchmark instances and that a substantial proportion of them can be solved to optimality in presolving -- without search. While we consider this result very promising, we do not observe significant change in overall run-time from the use of our novel dual reductions.

This paper discusses how to build a solver for mixed integer quadratically constrained programs (MIQCPs) by extending a framework for constraint integer programming (CIP). The advantage of this approach is that we can utilize the full power of advanced MIP and CP technologies. In particular, this addresses the linear relaxation and the discrete components of the problem. For relaxation, we use an outer approximation generated by linearization of convex constraints and linear underestimation of nonconvex constraints. Further, we give an overview of the reformulation, separation, and propagation techniques that are used to handle the quadratic constraints efficiently. We implemented these methods in the branch-cut-and-price framework SCIP. Computational experiments indicates the potential of the approach.

Pseudo-Boolean problems lie on the border between satisfiability problems, constraint programming, and integer programming. In particular, nonlinear constraints in pseudo-Boolean optimization can be handled by methods arising in these different fields: One can either linearize them and work on a linear programming relaxation or one can treat them directly by propagation. In this paper, we investigate the individual strengths of these approaches and compare their computational performance. Furthermore, we integrate these techniques into a branch-and-cut-and-propagate framework, resulting in an efficient nonlinear pseudo-Boolean solver.