Local Search Approach Research Articles

Bounded, Multidimensional, Integrated Memetic Evolution for Character Recognition Based on Predictive Elimination Theory and Optimization Techniques

This article describes how inspired by the natural process of evolution in genetic algorithms, memetic algorithms (MAs) are a category of cultural evolution phenomenon. The very concept of MA has been discussed in the last few years and is adding newer dimensions to MA and computational skills of algorithms. There are many optimization algorithms which fully exploit the problem under consideration. This article presents a heuristic approach for an improvised algorithm which takes into consideration various optimization parameters in isolation and tries to integrate the self-learning technique of MA. A general structure of MA according to this article should be perfectly in-line with brain activities which are neurotically tested and given maximum emphasis on local search and context-based predictive approaches rather than mathematically computing every event and taking or picking solutions based on results of certain formula. This article goes one step beyond the conventional set of the variety of problem domains, ranging from discrete optimization, continuous optimization, constrained optimization and multi objective optimization in which MAs have been successfully implemented. These optimization techniques must be processed using outcomes of predictive optimization and using a method of elimination to make the search set smaller and smaller as we progress deeper into the search. There is a scarcity of literature and also lack of availability of comprehensive reviews on MAs. The proposed technique is a better approach for solving combinatorial optimization problems. This article gives an overview of various domains and problem types in which MA can be used. Apart from this, the problem of character recognition using predictive optimization and implementation of elimination theory MA is discussed.

Two metaheuristics for solving no-wait operating room surgery scheduling problem under various resource constraints

The problem studied in this paper is operating room surgery scheduling, with resource constraints in each of the three following stages: preoperative, intraoperative, and postoperative stages. The availability of material resources, specialties and qualifications of human resources are integrated, and the aim is to schedule surgeries while minimizing the maximum end time of last activity in stage 3 and the total idle time in the operating rooms. Two metaheuristics, an iterative local search approach and a hybrid genetic algorithm, are provided and tested on real workday instances from the literature. Computational experiments showed that our metaheuristics outperformed the current state-of-the-art solving algorithm which is an ant colony optimization. The hybrid genetic algorithm reached small superiority vs. the iterative local search algorithm. The average reduction in the end time (the total idle time) was 24% (59%) with the iterated local search approach and 24% (70%) with the hybrid genetic algorithm vs. 14% (55%) with the ant colony optimization algorithm.

An Improved Multi-Objective Particle Swarm Optimization Based on Utopia Point Guided Search

This article demonstrates the implementation of a novel local search approach based on Utopia point guided search, thus improving the exploration ability of multi- objective Particle Swarm Optimization. This strategy searches for best particles based on the criteria of seeking solutions closer to the Utopia point, thus improving the convergence to the Pareto-optimal front. The elite non-dominated selected particles are stored in an archive and updated at every iteration based on least crowding distance criteria. The leader is chosen among the candidates in the archive using the same guided search. From the simulation results based on many benchmark tests, the new algorithm gives better convergence and diversity when compared to existing several algorithms such as NSGA-II, CMOPSO, SMPSO, PSNS, DE+MOPSO and AMALGAM. Finally, the proposed algorithm is used to solve mechanical design based multi-objective optimization problems from the literature, where it shows the same advantages.

A Cuckoo Search Algorithm to Solve Transfer Line Balancing Problems With Different Cutting Conditions

An automated transfer line balancing problem is investigated in this paper. The line produces a complex part at a high volume, such as cylinder heads, and incorporates identical machines that can be operated under different cutting conditions. The selection of cutting conditions significantly influences productivity and the cost of the transfer line. The production of cylinder heads requires machining different operations within a given takt time and satisfying precedence, inclusion, and exclusion relationships. The operations are located on different faces of the cylinder head and are processed by different cutting tools. The line balancing problem is studied to identify the optimal cutting conditions, number of machines and tools, and machining sequence of operations. The objective is to design a balanced transfer line at a minimum machining and tooling cost, and also with minimum nonproductive time. The problem is represented by a goal-programming model and solved for small transfer line configurations through linearization. An improved cuckoo search algorithm via Levy flight is developed to solve large-scale instances. Results of the cuckoo algorithm are promising since it reached optimal and close-to-optimal solutions to small problems and surpassed the results of a random local search approach for instances of medium and large problems.

A Grid Weighted Sum Pareto Local Search for Combinatorial Multi and Many-Objective Optimization.

Combinatorial multiobjective optimization problems (CMOPs) are very popular due to their widespread applications in the real world. One common method for CMOPs is Pareto local search (PLS), a natural extension of single-objective local search (LS). However, classical PLS tends to reserve all of the nondominated solutions for LS, which causes the inefficient LS, as well as unbearable computational and space cost. Due to the aforementioned reasons, most PLS approaches can only handle CMOPs with no more than two objectives. In this paper, by combining the Pareto dominance and weighted sum (WS) approach in a grid system, the grid weighted sum dominance (gws-dominance) is proposed and integrated into PLS for CMOPs with multiple objectives. In the grid system, at most one representative solution is maintained in each grid for more efficient LS, thus largely reducing the computational and space complexity. The grid-based WS approach can further guide the LS in different grids for maintaining more widely and uniformly distributed Pareto front approximations. In the experimental studies, the grid WS PLS is compared with the classical PLS, three decomposition-based LS approaches [multiobjective evolutionary algorithm based on decomposition-LS (WS, Tchebycheff, and penalty-based boundary intersection)], a grid-based algorithm ( ϵ -MOEA), and a state-of-the-art hybrid approach (multiobjective memetic algorithm based on decomposition) on two sets of benchmark CMOPs. The experimental results show that the grid weighted sum Pareto local search significantly outperforms the compared algorithms and remains effective and efficient on combinatorial multiobjective and even many-objective optimization problems.

A Local Search Approach to Observation Planning with Multiple UAVs

Observation planning for Unmanned Aerial Vehicles (UAVs) is a challenging task as it requires planning trajectories over a large continuous space and with motion models that can not be directly encoded into current planners. Furthermore, realistic problems often require complex objective functions that complicate problem decomposition. In this paper, we propose a local search approach to plan the trajectories of a fleet of UAVs on an observation mission. The strength of the approach lies in its loose coupling with domain specific requirements such as the UAV model or the objective function that are both used as black boxes. Furthermore, the Variable Neighborhood Search (VNS) procedure considered facilitates the adaptation of the algorithm to specific requirements through the addition of new neighborhoods. We demonstrate the feasibility and convenience of the method on a large joint observation task in which a fleet of fixed-wing UAVs maps wildfires over areas of a hundred square kilometers. The approach allows generating plans over tens of minutes for a handful of UAVs in matter of seconds, even when considering very short primitive maneuvers.

Finding optimal mentor-mentee matches: A case study in applied two-sided matching

Two-Sided Matching is a well-established approach to find allocations and matchings based on the participants' preferences. While its most prominent applications are College Admissions and School Choice problems, this paper applies the concept to the matching of mentors to mentees in a higher education context. Both mentors and mentees have preferences with whom they ideally want to be matched, as well as who they want to avoid. As the general formulation for these types of preferences is NP-hard, several existing approximation algorithms and heuristics are compared with respect to their ability to find a matching with desirable properties. The results show that a combination of evolutionary heuristics and local search approaches works best in finding high-quality solutions, allowing us to find mentor-mentee pairs which are close to the respective ideal match.

Randomized Decomposition Solver with the Quadratic Assignment Problem as a Case Study

This paper presents a new local search approach, called randomized decomposition (RD), for solving nonlinear, nonconvex mathematical programs. Starting from a feasible solution, RD partitions the problem’s decision variables into a randomly ordered list of randomly generated subsets. RD then optimizes over the variables in each subset, keeping all other variables fixed. Unlike most other decomposition methods, no knowledge of the problem structure is required. RD has been combined with a metaheuristic RDPerturb, for escaping local optima, to create a generic framework for solving mathematical programs, especially hard combinatorial nonconvex problems. The framework has been implemented as an optimization platform we call RDSolver and successfully applied to over 400 instances of the quadratic assignment problem (QAP). The results obtained by RDSolver are competitive with the solutions obtained by heuristics specially tailored for those problems, even though RDSolver is a general purpose mathematical progr...

Tabu search and iterated local search for the cyclic bottleneck assignment problem

In this paper, we study a new variant of the classical assignment problem referred to as the cyclic bottleneck assignment problem (CBAP). This problem arises in surgical and operating room (OR) scheduling and can be described as follows. Each of n successive block times can be allocated to anyone of n surgeons. The number of patients to be scheduled for surgery on a given block and the length of stay in the recovery unit of each operated patient are dependent on the type of surgery. The problem is to assign one surgeon to each OR time so as to minimize the maximum number of patients that will be expected to occupy the recovery unit in the long run.This problem has been shown to be NP-hard. Consequently, we present two algorithms that can be used to solve the problem efficiently. One algorithm is based on the Tabu search method, and the other follows an iterated local search scheme. We evaluate our algorithms on 470 benchmark instances and 470 newly generated instances, and compare them with available metaheuristics and commercial solver CPLEX. The computational results for most of our problem instances show that our proposed approaches achieve excellent performance. Our algorithms outperform existing approaches and CPLEX. The tabu search approach works slightly better than the iterated local search approach.

Bi-level hybrid local search approach for three-dimensional loading problem with balancing constraints

This paper presents a bi-level hybrid local search (BHLS) algorithm for the three-dimensional loading problem with balancing constraints (3DLP-B), where several rectangular boxes with even densities but different sizes are loaded into a single cubic bin to meet the requirements of the space or capacity utilization and the balance of the center of gravity. The proposed algorithm hybridizes a novel framed-layout procedure in which the concept of the core block and its generation strategy are introduced. Once the block-loading sequence has been determined, we can load one block at a time by the designed construction heuristic. Then, the double-search is introduced; its external search procedure generates a list of compact packing patterns while its internal search procedure is used to search the core-block frames and their best distribution locations. The approach is extensively tested on weakly to strongly heterogeneous benchmark data. The results show that it has better performance in improving space utilization rate and balanced condition of the placement than existed techniques: the overall averages from 79.85% to 86.45% were obtained for the balanced cases and relatively high space-usage rate of 89.44% was achieved for the unbalanced ones.

A memetic differential evolution algorithm for energy-efficient parallel machine scheduling

This paper considers an energy-efficient bi-objective unrelated parallel machine scheduling problem to minimize both makespan and total energy consumption. The parallel machines are speed-scaling. To solve the problem, we propose a memetic differential evolution (MDE) algorithm. Since the problem involves assigning jobs to machines and selecting an appropriate processing speed level for each job, we characterize each individual by two vectors: a job-machine assignment vector and a speed vector. To accelerate the convergence of the algorithm, only the speed vector of each individual evolves and a list scheduling heuristic is applied to derive its job-machine assignment vector based on its speed vector. To further enhance the algorithm, we propose efficient speed adjusting and job-machine swap heuristics and integrate them into the algorithm as a local search approach by an adaptive meta-Lamarckian learning strategy. Computational results reveal that the incorporation of list scheduling heuristic and local search greatly strengthens the algorithm. Computational experiments also show that the proposed MDE algorithm outperforms SPEA-II and NSGA-II significantly.

Learning compact zero-order TSK fuzzy rule-based systems for high-dimensional problems using an Apriori + local search approach

Learning fuzzy rule-based systems entails searching a set of fuzzy rules which fits the training data. Even if using fix fuzzy partitions, the amount of rules that can be formed is exponential in the number of variables. Thus, the search must be carried out by means of metaheuristics such as genetic algorithms, and sometimes restricted to the set of candidate rules which reach a minimum support.In this article, we propose and evaluate two methods to learn zero-order Takagi–Sugeno–Kang fuzzy systems in medium-high dimensional domains. First, we introduce the minimum individual error of a rule in the criterion for candidate selection. Then, due to the intrinsic locality of fuzzy rule-based systems, where each rule mainly interacts with adjacent rules, we study the use of local search algorithms to carry out the search of the final rule base.Results show that the proposed scheme for candidate rule selection leads to an improvement in the results, regardless of the subsequent search algorithm. Moreover, local search-based algorithms achieve competitive results, reducing substantially the number of rules of the learnt systems.

An efficient guided local search approach for multiuser detection in UWB systems

In this paper, we tackled the performance of a multiuser detection (MUD) in ultra wideband (UWB) systems over extreme multipath channel model (CM4) based on TH-PPM and DS-PAM in a rake receiver. For better performance at receiver side, a novel guided local search (GLS) algorithm is proposed for UWB systems. Simulation results show a significant bit error rate (BER) performance gain can be achieved based on the proposed GLS-MUD method, with conventional detector (CD). In addition, the technique is very promising when the number of interfering users increases.

Propagation based local search for bit-precise reasoning

Many applications of computer-aided verification require bit-precise reasoning as provided by satisfiability modulo theories (SMT) solvers for the theory of quantifier-free fixed-size bit-vectors. The current state-of-the-art in solving bit-vector formulas in SMT relies on bit-blasting, where a given formula is eagerly translated into propositional logic (SAT) and handed to an underlying SAT solver. Bit-blasting is efficient in practice, but may not scale if the input size can not be reduced sufficiently during preprocessing. A recent score-based local search approach lifts stochastic local search from the bit-level (SAT) to the word-level (SMT) without bit-blasting and proved to be quite effective on hard satisfiable instances, particularly in the context of symbolic execution. However, it still relies on brute-force randomization and restarts to achieve completeness. Guided by a completeness proof, we simplified, extended and formalized our propagation-based variant of this approach. We obtained a clean, simple and more precise algorithm that does not rely on score-based local search techniques and does not require brute-force randomization or restarts to achieve completeness. It further yields substantial gain in performance. In this article, we present and discuss our complete propagation based local search approach for bit-vector logics in SMT in detail. We further provide an extended and extensive experimental evaluation including an analysis of randomization effects.

Consolidated optimization algorithm for resource-constrained project scheduling problems

Resource-constrained project scheduling problems (RCPSPs) represent an important class of practical problems. Over the years, many optimization algorithms for solving them have been proposed, with their performances evaluated using well-established test instances with various levels of complexity. While it is desirable to obtain a high-quality solution and fast rate of convergence from an optimization algorithm, no single one performs well across the entire space of instances. Furthermore, even for a given algorithm, the optimal choice of its operators and control parameters may vary from one problem to another. To deal with this issue, we present a generic framework for solving RCPSPs in which various meta-heuristics, each with multiple search operators, are self-adaptively used during the search process and more emphasis is placed on the better-performing algorithms, and their underlying search operators. To further improve the rate of convergence and introduce good-quality solutions into the population earlier, a local search approach is introduced. The experimental results clearly indicate the capability of the proposed algorithm to attain high-quality results using a small population. Compared with several state-of-the-art algorithms, the proposed one delivers the best solutions for problems with 30 and 60 activities, and is very competitive for those involving 120 activities.

Approaches for Job Shop Scheduling in Swarm techniques

The behaviour of a single ant, bee, termite and wasp often is too simple, but their collective and social behaviour is of paramount significance. The National Geographic TV Channel reveals that advanced mammals including lions also enjoy social lives, perhaps for their self-existence at old age and in particular when they are wounded. Hence, most optimization approaches are only suitable for small problems practically. The local search approaches are all based on neighbourhood structures defining how to obtain a new solution from the current ones, such as tabu search and simulated algorithm. In this paper, we develop an ant colony optimization approach to the job shop scheduling problem that utilizes a local search technique.

A constraint-based parallel local search for the edge-disjoint rooted distance-constrained minimum spanning tree problem

Many network design problems arising in areas as diverse as VLSI circuit design, QoS routing, traffic engineering, and computational sustainability require clients to be connected to a facility under path-length constraints and budget limits. These problems can be seen as instances of the rooted distance-constrained minimum spanning-tree problem (RDCMST), which is NP-hard. An inherent feature of these networks is that they are vulnerable to a failure. Therefore, it is often important to ensure that all clients are connected to two or more facilities via edge-disjoint paths. We call this problem the edge-disjoint RDCMST (ERDCMST). Previous work on the RDCMST has focused on dedicated algorithms and therefore it is difficult to use these algorithms to tackle the ERDCMST. We present a constraint-based parallel local search algorithm for solving the ERDCMST. Traditional ways of extending a sequential algorithm to run in parallel perform either portfolio-based search in parallel or parallel neighbourhood search. Instead, we exploit the semantics of the constraints of the problem to perform multiple moves in parallel by ensuring that they are mutually independent. The ideas presented in this paper are general and can be adapted to other problems as well. The effectiveness of our approach is demonstrated by experimenting with a set of problem instances taken from real-world passive optical network deployments in Ireland, Italy, and the UK. Our results show that performing moves in parallel can significantly reduce the elapsed time and improve the quality of the solutions of our local search approach.

An Improved Multi-Objective Particle Swarm Optimization Algorithm Based on Adaptive Local Search

This paper demonstrates a novel local search approach based on an adaptive (time variant) search space index improving the exploration ability as well as diversity in multi-objective Particle Swarm Optimization. The novel strategy searches for the neighbourhood particles in a range which gradually increases with iterations. Particles get updated according to the rules of basic PSO and the non-dominated particles are subjected to Evolutionary update archiving. To improve the diversity, the archive is truncated based on crowding distance parameter. The leader is chosen among the candidates in the archive based on another local search. From the simulation results, it is clear that the implementation of the new scheme results in better convergence and diversity as compared to NSGA-II, CMPSO, and SMPSO reported in literature. Finally, the proposed algorithm is used to solve machine design based engineering problems from literature and compared with existing algorithms.

A local search enhanced differential evolutionary algorithm for sparse recovery

Signal recovery problem in compressed sensing can be mathematically modeled as a ℓ0 regularized problem, which aims at searching a sparse solution. When tackling this problem, traditional mathematical approaches suffer from a limited convergence ability, especially under the noisy condition. To better solve this problem, in this paper, a novel differential evolutionary (DE) algorithm is designed to combine with a local search approach. First, an adaptive control strategy for DE is extended to recover sparse signals with noise in this paper, which is found to have a promising recovery performance. Second, in order to further enhance the convergence speed, a local search approach, i.e., a shrinkage-thresholding method (STM), is embedded into the evolutionary process of DE. Therefore, the advantages of local search capability provided by STM and global search ability of DE can be effectively combined, and resultantly a novel local search enhanced adaptive DE (named LSE-ADE) algorithm is proposed. Experimental results validate that LSE-ADE performs better than the eight classic sparse recovery algorithms and one recently proposed evolutionary algorithm, when recovering sparse signal under the noisy condition.

Order Batching and Picker Routing in manual order picking systems: the benefits of integrated routing

Order Batching and Picker Routing Problems arise in warehouses when items specified by customer orders have to be retrieved from their storage locations. The Order Batching Problem includes the grouping of a given set of customer orders into feasible picking orders such that the total length of all picker tours is minimized. In order to calculate the length of a picker tour, the sequence has to be determined according to which the items contained in the picking order will be picked. This problem is known as the Picker Routing Problem. Although quite sophisticated heuristics and even efficient exact solution approaches exist to the Picker Routing Problem in warehouse with up to two blocks, the routing problem does not get much attention when dealing with the Order Batching Problem. Instead, the order pickers are assumed to follow a certain, simple routing strategy when making their ways through the warehouse. The advantage of this approach can be seen in the fact that—in particular for single-block warehouse layouts—the corresponding picker tours are very straightforward and can be memorized easily by the order pickers. This advantage diminishes, however, when more complex, multi-block layouts have to be dealt with. Furthermore, in such case, the approach may result in picker tours which are far from optimal. For multi-block layouts, we integrate different routing algorithms into an iterated local search approach for the batching in order to demonstrate what the benefits are from solving the Order Batching and the Picker Routing Problem in a more integrated way. By means of numerical experiments it is shown that paying more attention to the Picker Routing Problem results in a substantial improvement of the solution quality without increasing computing times.