Convex Mixed-integer Nonlinear Programming Research Articles

An Outer-Inner Approximation for Separable Mixed-Integer Nonlinear Programs

A common structure in convex mixed-integer nonlinear programs (MINLPs) is separable nonlinear functions. In the presence of such structures, we propose three improvements to the outer approximation algorithms. The first improvement is a simple extended formulation, the second is a refined outer approximation, and the third is a heuristic inner approximation of the feasible region. As a side result, we exhibit a simple example where a classical implementation of the outer approximation would take an exponential number of iterations, whereas it is easily solved with our modifications. These methods have been implemented in the open source solver Bonmin and are available for download from the Computational Infrastructure for Operations Research project website. We test the effectiveness of the approach on three real-world applications and on a larger set of models from an MINLP benchmark library. Finally, we show how the techniques can be extended to perspective formulations of several problems. The proposed tools lead to an important reduction in computing time on most tested instances.

On MINLP Heuristics for Solving Shale-well Scheduling Problems

This paper describes mixed integer nonlinear programming (MINLP) heuristics for solving dynamic scheduling problems in complex petroleum production systems with a network topology. We modify the Feasibility Pump heuristic for convex MINLPs [Bonami and Gonçalves, 2010] by formulating a multiobjective problem, in which we aim at balancing the two goals of quickly obtaining a feasible solution and preserving solution quality with respect to the objective value. We further present a simple linearization-based heuristic, only aimed at quickly generating feasible solutions. The MINLP heuristics are applied to a dynamic multi-pipeline shale well and compressor scheduling problem, targeted on application in decision-support tools for improving operations in large shale-gas systems. Developing efficient and robust heuristics are important for the applicability of these tools, in the sense that low computation times are often more important than global optima. A computational study shows that the proposed objective-oriented Feasibility Pump is competitive both in terms of solution quality and computation time compared to other heuristics and the branch-and-bound method.

On branching rules for convex mixed-integer nonlinear optimization

Branch-and-Bound (B&B) is perhaps the most fundamental algorithm for the global solution of convex Mixed-Integer Nonlinear Programming (MINLP) problems. It is well-known that carrying out branching in a nonsimplistic manner can greatly enhance the practicality of B&B in the context of Mixed-Integer Linear Programming (MILP). No detailed study of branching has heretofore been carried out for MINLP. In this article, we study and identify useful sophisticated branching methods for MINLP, including novel approaches based on approximations of the nonlinear relaxations by linear and quadratic programs.

A deterministic global approach for mixed-discrete structural optimization

This study proposes a novel approach for finding the exact global optimum of a mixed-discrete structural optimization problem. Although many approaches have been developed to solve the mixed-discrete structural optimization problem, they cannot guarantee finding a global solution or they adopt too many extra binary variables and constraints in reformulating the problem. The proposed deterministic method uses convexification strategies and linearization techniques to convert a structural optimization problem into a convex mixed-integer nonlinear programming problem solvable to obtain a global optimum. To enhance the computational efficiency in treating complicated problems, the range reduction technique is also applied to tighten variable bounds. Several numerical experiments drawn from practical structural design problems are presented to demonstrate the effectiveness of the proposed method.

Mixed-integer nonlinear optimization

Many optimal decision problems in scientific, engineering, and public sector applications involve both discrete decisions and nonlinear system dynamics that affect the quality of the final design or plan. These decision problems lead to mixed-integer nonlinear programming (MINLP) problems that combine the combinatorial difficulty of optimizing over discrete variable sets with the challenges of handling nonlinear functions. We review models and applications of MINLP, and survey the state of the art in methods for solving this challenging class of problems.Most solution methods for MINLP apply some form of tree search. We distinguish two broad classes of methods: single-tree and multitree methods. We discuss these two classes of methods first in the case where the underlying problem functions are convex. Classical single-tree methods include nonlinear branch-and-bound and branch-and-cut methods, while classical multitree methods include outer approximation and Benders decomposition. The most efficient class of methods for convex MINLP are hybrid methods that combine the strengths of both classes of classical techniques.Non-convex MINLPs pose additional challenges, because they contain non-convex functions in the objective function or the constraints; hence even when the integer variables are relaxed to be continuous, the feasible region is generally non-convex, resulting in many local minima. We discuss a range of approaches for tackling this challenging class of problems, including piecewise linear approximations, generic strategies for obtaining convex relaxations for non-convex functions, spatial branch-and-bound methods, and a small sample of techniques that exploit particular types of non-convex structures to obtain improved convex relaxations.We finish our survey with a brief discussion of three important aspects of MINLP. First, we review heuristic techniques that can obtain good feasible solution in situations where the search-tree has grown too large or we require real-time solutions. Second, we describe an emerging area of mixed-integer optimal control that adds systems of ordinary differential equations to MINLP. Third, we survey the state of the art in software for MINLP.

Design Optimization of a Speed Reducer Using Deterministic Techniques

The optimal design problem of minimizing the total weight of a speed reducer under constraints is a generalized geometric programming problem. Since the metaheuristic approaches cannot guarantee to find the global optimum of a generalized geometric programming problem, this paper applies an efficient deterministic approach to globally solve speed reducer design problems. The original problem is converted by variable transformations and piecewise linearization techniques. The reformulated problem is a convex mixed-integer nonlinear programming problem solvable to reach an approximate global solution within an acceptable error. Experiment results from solving a practical speed reducer design problem indicate that this study obtains a better solution comparing with the other existing methods.

Inexact solution of NLP subproblems in MINLP

In the context of convex mixed integer nonlinear programming (MINLP), we investigate how the outer approximation method and the generalized Benders decomposition method are affected when the respective nonlinear programming (NLP) subproblems are solved inexactly. We show that the cuts in the corresponding master problems can be changed to incorporate the inexact residuals, still rendering equivalence and finiteness in the limit case. Some numerical results will be presented to illustrate the behavior of the methods under NLP subproblem inexactness.

A storm of feasibility pumps for nonconvex MINLP

One of the foremost difficulties in solving Mixed-Integer Nonlinear Programs, either with exact or heuristic methods, is to find a feasible point. We address this issue with a new feasibility pump algorithm tailored for nonconvex Mixed-Integer Nonlinear Programs. Feasibility pumps are algorithms that iterate between solving a continuous relaxation and a mixed-integer relaxation of the original problems. Such approaches currently exist in the literature for Mixed-Integer Linear Programs and convex Mixed-Integer Nonlinear Programs: both cases exhibit the distinctive property that the continuous relaxation can be solved in polynomial time. In nonconvex Mixed-Integer Nonlinear Programming such a property does not hold, and therefore special care has to be exercised in order to allow feasibility pump algorithms to rely only on local optima of the continuous relaxation. Based on a new, high level view of feasibility pump algorithms as a special case of the well-known successive projection method, we show that many possible different variants of the approach can be developed, depending on how several different (orthogonal) implementation choices are taken. A remarkable twist of feasibility pump algorithms is that, unlike most previous successive projection methods from the literature, projection is “naturally” taken in two different norms in the two different subproblems. To cope with this issue while retaining the local convergence properties of standard successive projection methods we propose the introduction of appropriate norm constraints in the subproblems; these actually seem to significantly improve the practical performance of the approach. We present extensive computational results on the MINLPLib, showing the effectiveness and efficiency of our algorithm.

A reformulation framework for global optimization

In this paper, we present a global optimization method for solving nonconvex mixed integer nonlinear programming (MINLP) problems. A convex overestimation of the feasible region is obtained by replacing the nonconvex constraint functions with convex underestimators. For signomial functions single-variable power and exponential transformations are used to obtain the convex underestimators. For more general nonconvex functions two versions of the so-called ?BB-underestimator, valid for twice-differentiable functions, are integrated in the actual reformulation framework. However, in contrast to what is done in branch-and-bound type algorithms, no direct branching is performed in the actual algorithm. Instead a piecewise convex reformulation is used to convexify the entire problem in an extended variable-space, and the reformulated problem is then solved by a convex MINLP solver. As the piecewise linear approximations are made finer, the solution to the convexified and overestimated problem will form a converging sequence towards a global optimal solution. The result is an easily-implementable algorithm for solving a very general class of optimization problems.

An alternative strategy for global optimization of heat exchanger networks

The HEN synthesis problem is one amongst many engineering problems which can be characterized as highly combinatorial, nonlinear and nonconvex, all contributing to computational difficulties shown either in a form of long computational times and/or in identifying poor locally optimal solutions. In this work, a new strategy for global optimization of heat exchanger networks (HENs) is presented. We first introduce a concept of stage-wise superstructure augmented by an aggregated substructure. On this basis, the HEN synthesis problem is formulated as a mixed integer nonlinear program (MINLP). The strategy for providing globally optimal solutions relies on solving a single convex MINLP which incorporates piecewise linear and nonlinear convex underestimators of the nonconvex linear fractional terms present in the nonconvex MINLP. It is shown that the optimal solution of the convex MINLP can provide a lower bound tight enough that the gap between the upper and lower bound falls below 1%. In addition, an algorithm for identifying good locally optimal solutions is presented. The approach was tested on two examples, showing that currently we are able to solve small HEN synthesis problems to global optimality with reasonable computational effort, while good locally optimal solutions can be identified for larger problems.

Optimization of Supply Chain Systems with Price Elasticity of Demand

A centralized multiechelon, multiproduct supply chain network is presented in a multiperiod setting with products that show varying demand against price. An important consideration in such complex supply chains is to maintain system performance at high levels for varying demands that may be sensitive to product price. To examine the price-centric behavior of the customers, the concept of price elasticity of demand is addressed. The proposed approach includes many realistic features of typical supply chain systems such as production planning and scheduling, inventory management, transportation delay, transportation cost, and transportation limits. In addition, the proposed system can be extended to meet unsatisfied demand in future periods by backordering. Effects of the elasticity in price demand in production and inventory decisions are also examined. The supply chain model is formulated as a convex mixed-integer nonlinear programming problem. Reformulations are presented to make the problem tractable. The differential equations are reformulated as difference equations, and unbounded derivatives in the nonlinear objective function are handled with an approximation, with guaranteed bounds on the loss of optimality. The approach is illustrated on a multiechelon, multiproduct supply chain network.

An Efficient Global Approach for Posynomial Geometric Programming Problems

Aposynomial geometric programming problem is composed of a posynomial being minimized in the objective function subject to posynomial constraints. This study proposes an efficient method to solve a posynomial geometric program with separable functions. Power transformations and exponential transformations are utilized to convexify and underestimate posynomial terms. The inverse transformation functions of decision variables generated in the convexification process are approximated by superior piecewise linear functions. The original program therefore can be converted into a convex mixed-integer nonlinear program solvable to obtain a global optimum. Several numerical experiments are presented to investigate the impact of different convexification strategies on the obtained approximate solution and to demonstrate the advantages of the proposed method in terms of both computational efficiency and solution quality.

Range reduction techniques for improving computational efficiency in global optimization of signomial geometric programming problems

Many global optimization approaches for solving signomial geometric programming problems are based on transformation techniques and piecewise linear approximations of the inverse transformations. Since using numerous break points in the linearization process leads to a significant increase in the computational burden for solving the reformulated problem, this study integrates the range reduction techniques in a global optimization algorithm for signomial geometric programming to improve computational efficiency. In the proposed algorithm, the non-convex geometric programming problem is first converted into a convex mixed-integer nonlinear programming problem by convexification and piecewise linearization techniques. Then, an optimization-based approach is used to reduce the range of each variable. Tightening variable bounds iteratively allows the proposed method to reach an approximate solution within an acceptable error by using fewer break points in the linearization process, therefore decreasing the required CPU time. Several numerical experiments are presented to demonstrate the advantages of the proposed method in terms of both computational efficiency and solution quality.

Finding multiple optimal solutions of signomial discrete programming problems with free variables

With the increasing reliance on mathematical programming based approaches in various fields, signomial discrete programming (SDP) problems occur frequently in real applications. Since free variables are often introduced to model problems and alternative optima are practical for decision making among multiple strategies, this paper proposes a generalized method to find multiple optimal solutions of SDP problems with free variables. By means of variable substitution and convexification strategies, an SDP problem with free variables is first converted into another convex mixed-integer nonlinear programming problem solvable to obtain an exactly global optimum. Then a general cut is utilized to exclude the previous solution and an algorithm is developed to locate all alternative optimal solutions. Finally, several illustrative examples are presented to demonstrate the effectiveness of the proposed approach.

FilMINT: An Outer Approximation-Based Solver for Convex Mixed-Integer Nonlinear Programs

We describe a new solver for convex mixed-integer nonlinear programs (MINLPs) that implements a linearization-based algorithm. The solver is based on an algorithm of Quesada and Grossmann [Quesada, I., I. E. Grossmann. 1992. An LP/NLP based branch-and-bound algorithm for convex MINLP optimization problems. Comput. Chemical Engrg. 16(10–11) 937–947] that avoids the complete re-solution of a master mixed-integer linear program (MILP) by adding new linearizations at open nodes of the branch-and-bound tree whenever an integer solution is found. The new solver, FilMINT, combines the MINTO branch-and-cut framework for MILP with filterSQP to solve the nonlinear programs that arise as subproblems in the algorithm. The MINTO framework allows us to easily employ cutting planes, primal heuristics, and other well-known MILP enhancements for MINLPs. We present detailed computational experiments that show the benefit of such advanced MILP techniques. We offer new suggestions for generating and managing linearizations that are shown to be efficient on a wide range of MINLPs. By carefully incorporating and tuning all these enhancements, an effective solver for convex MINLPs is constructed.

Route optimization for multiple searchers

AbstractWe consider a discrete time‐and‐space route‐optimization problem across a finite time horizon in which multiple searchers seek to detect one or more probabilistically moving targets. This article formulates a novel convex mixed‐integer nonlinear program for this problem that generalizes earlier models to situations with multiple targets, searcher deconfliction, and target‐ and location‐dependent search effectiveness. We present two solution approaches, one based on the cutting‐plane method and the other on linearization. These approaches result in the first practical exact algorithms for solving this important problem, which arises broadly in military, rescue, law enforcement, and border patrol operations. The cutting‐plane approach solves many realistically sized problem instances in a few minutes, while existing branch‐and‐bound algorithms fail. A specialized cut improves solution time by 50[percnt] in difficult problem instances. The approach based on linearization, which is applicable in important special cases, may further reduce solution time with one or two orders of magnitude. The solution time for the cutting‐plane approach tends to remain constant as the number of searchers grows. In part, then, we overcome the difficulty that earlier solution methods have with many searchers. © 2010 Wiley Periodicals, Inc. Naval Research Logistics, 2010

Heuristics for convex mixed integer nonlinear programs

In this paper, we describe the implementation of some heuristics for convex mixed integer nonlinear programs. The work focuses on three families of heuristics that have been successfully used for mixed integer linear programs: diving heuristics, the Feasibility Pump, and Relaxation Induced Neighborhood Search (RINS). We show how these heuristics can be adapted in the context of mixed integer nonlinear programming. We present results from computational experiments on a set of instances that show how the heuristics implemented help finding feasible solutions faster than the traditional branch-and-bound algorithm and how they help in reducing the total solution time of the branch-and-bound algorithm.

A Feasibility Pump for mixed integer nonlinear programs

We present an algorithm for finding a feasible solution to a convex mixed integer nonlinear program. This algorithm, called Feasibility Pump, alternates between solving nonlinear programs and mixed integer linear programs. We also discuss how the algorithm can be iterated so as to improve the first solution it finds, as well as its integration within an outer approximation scheme. We report computational results.

Global solution of optimization problems with signomial parts

In this paper a new approach for the global solution of nonconvex MINLP (Mixed Integer NonLinear Programming) problems that contain signomial (generalized geometric) expressions is proposed and illustrated. By applying different variable transformation techniques and a discretization scheme a lower bounding convex MINLP problem can be derived. The convexified MINLP problem can be solved with standard methods. The key element in this approach is that all transformations are applied termwise. In this way all convex parts of the problem are left unaffected by the transformations. The method is illustrated by four example problems.

Solving Convex MINLP Optimization Problems Using a Sequential Cutting Plane Algorithm

In this article we look at a new algorithm for solving convex mixed integer nonlinear programming problems. The algorithm uses an integrated approach, where a branch and bound strategy is mixed wit...