Search

This paper studies the customer order scheduling problem in the context of additive manufacturing. The study discusses an integrated problem involving the nesting of parts as well as the scheduling of batches of nested parts onto unrelated parallel machines. A mixed-integer programming model is presented, based on existing formulations from the literature, that integrates different materials and sequence-dependent setup times. Additionally, a metaheuristic based on an iterated local search is proposed for the problem configuration under consideration. Focusing on minimizing the total weighted tardiness of orders, the efficiency of the heuristic approach is evaluated using comprehensive test data. Further, we show the importance of the considered order-related objective by using qualitative analysis. [ABSTRACT FROM AUTHOR]

A scheduling problem for a two‐stage flexible flow shop with s‐batching machines motivated by processes in additive manufacturing is considered. A batch is a group of jobs that are processed together on a single machine. Each job belongs to an incompatible family. Only jobs of the same family can be batched together. A maximum batch size is given. A setup time occurs between different batches. The jobs of a batch are processed in a serial manner, that is, the processing time of a batch is the sum of the processing times of the jobs forming the batch. Batch availability is assumed. A job has a weight, a due date, and a release date. The total weighted tardiness is considered as performance measure. We establish a mixed integer linear programming formulation for this scheduling problem. For large‐sized problem instances, an iterative decomposition approach (IDA) is proposed that uses a grouping genetic algorithm or an iterated local search (ILS) scheme to solve the single‐stage subproblems. Moreover, an alternative ILS scheme based on a disjunctive graph representation of the problem at hand is designed. Results of computational experiments based on randomly generated problem instances demonstrate that the IDA hybridized with ILS outperforms the two other schemes. [ABSTRACT FROM AUTHOR]

Sports timetabling is a complex and challenging problem. The latest open benchmark dataset for the sport timetabling problem is from the International Timetabling Competition (ITC) 2021. Due to its complexity, only a few approaches have successfully generated feasible solutions for the problems in this dataset, as reported in scientific literature. To the best of our knowledge, there is only one study in the literature that has successfully generated feasible solutions for all 45 problems in the dataset. In this paper, we propose our novel efficient algorithm based on the Iterated Local Search algorithm to solve the ITC 2021 benchmark dataset. Unlike prior successful approaches that combined metaheuristics with an exact approach, our proposed approach is solely metaheuristic. Our contribution includes the design of strategies for both perturbation and local search phases, coupled with the integration of shuffling strategies. The experimental results show that our proposed algorithm is remarkably successful in generating feasible solutions for all 45 problems present in the ITC 2021 dataset. [ABSTRACT FROM AUTHOR]

This work address a variant of the knapsack problem, known as the knapsack problem with forfeits, which has numerous applications. In this variant, a set of items and a conflict graph are given, and the objective is to identify a collection of items that adhere to the knapsack's capacity while maximizing the total value of the items minus the penalties for conflicting items. We propose a novel heuristic for this problem based on the concepts of iterated local search, variable neighborhood descent, and tabu search. Our heuristic takes into account four neighborhood structures, and we introduce efficient data structures to explore them. Experimental results demonstrate that our approach outperforms the state-of-the-art algorithms in the literature. In particular, it delivers superior solutions within significantly shorter computation times across all benchmark instances. Additionally, this study includes an analysis of how the proposed data structures have influenced both the quality of the solutions and the execution time of the method. [ABSTRACT FROM AUTHOR]

This paper presents a facility location problem with capacitated multiservice facilities in which the allocation of the users is based on their preferences. The problem arises from an educational program, where a government agency or a nongovernmental organization (NGO) offers extracurricular workshops to elementary school students. These workshops promote the development of academic, social, or cultural skills of the students. This problem is formulated as a bi‐level program. The government/NGO is associated with the upper level, which aims to minimize the costs for opening workshops in different schools and the costs of allocating students to each workshop. The lower level is related to the students and seeks to optimize their global preferences for the offered workshops. To solve this problem, we propose a single‐level reformulation and two metaheuristics (a path‐relinking metaheuristic (PRM) and an iterated local search (ILS) procedure). The best performing method is the PRM that generates saturated initial random solutions following a drop heuristic approach. After different phases of improvement (with drop and exchange heuristics), the solutions undergo a path‐relinking procedure. The trajectory between the initial solution and the best solution found so far is explored using a randomized backward path‐relinking strategy that thoroughly explores the neighborhood of the best solution. Computational experimentation indicates that the proposed PRM metaheuristic is effective and efficient. The results obtained provide better upper bounds in shorter running times than the ones obtained by the single‐level reformulation or the ILS procedure. Finally, the application to a realistic case study of Apodaca municipality in Nuevo Leon, Mexico, shows that large‐scale instances can be solved efficiently using the proposed metaheuristic. [ABSTRACT FROM AUTHOR]

This paper addresses the Hybrid Flow Shop Scheduling Problem (HFSSP) by integrating metaheuristic (MHs) and machine learning (ML) approaches. Specifically, we propose a hybrid algorithm by combining Ant Colony Optimization (ACO) and Iterated Local Search (ILS) to form ACOILS. To further enhance the performance of this hybrid approach, we employ Proximal Policy Optimization (PPO), which is used for dynamic tuning of key parameters within the hybrid algorithm. The introduction of PPO allows real-time adjustment of key parameters, such as pheromone evaporation rates and local search intensity, to balance exploration and exploitation more effectively. Comparative experiments against the non-learning version of ACOILS and Simulated Annealing (SA) show that the learning based LACOILS significantly reduces the percentage deviation from the lower bound while maintaining stable performance through dynamic tuning. In terms of numerical results, LACOILS consistently outperforms SA and ACOILS. For smaller instances (N=20), it achieves up to 56.52% improvement over ACOILS and 12.5% over SA. For larger instances (N=150), LACOILS shows up to 29.82% improvement over ACOILS and 9.09% over SA, demonstrating its superior solution quality and efficiency.

In this work, we introduce a generalisation of the Vehicle Routing Problem for a specific application in the monitoring of a Water Distribution Network (WDN). In this problem, multiple technicians must visit a sequence of nodes in the WDN and perform a series of tests to check the quality of water. Some special nodes (i.e., wells) require technicians to first collect a key from a key centre. The key must then be returned to the same key centre after the test has been performed, thus introducing precedence constraints and multiple visits in the routes. To solve the problem, a Mixed Integer Linear Programming model and an Iterated Local Search have been implemented. The efficiency of the proposed methods is demonstrated by means of extensive computational tests on randomly created and real-world instances. [ABSTRACT FROM AUTHOR]

The minimum multiplex arrangement problem locates each vertex of a graph on an integer point on a line, with each integer point being assigned at most $ \mathfrak {m} $ m vertices. The case with $ \mathfrak {m}=2 $ m = 2 , known as the minimum duplex arrangement (MinDA) problem, is investigated in this study. MinDA is $ \mathscr {N}\!\!\mathscr {P} $ N P -hard and is related to applications such as manufacturing systems design and VLSI layout. A mixed-integer linear programming formulation of MinDA is presented. The capabilities of the proposed formulation are assessed on 63 problem instances with varying densities and sizes up to 25 vertices. Additionally, an examination of the structure of an optimal MinDA solution is conducted. Based on this analysis, an iterated local search (ILS) algorithm is established, and novel strategies are developed to enhance its effectiveness. In contrast to traditional ILS approaches, where the local search phase invariably reaches a local optimum, the local search phase in the proposed algorithm may converge to a point in the vicinity of a local optimum, thereby reducing the computational burden for large-scale instances. Extensive computational experiments employing the ILS algorithm are conducted on instances with sizes n up to 10,240 vertices. [ABSTRACT FROM AUTHOR]

The Single Source Capacitated Facility Location Problem (SSCFLP) consists of determining locations for facilities to meet customer demands so that each customer must be served by a single facility. This paper proposes a matheuristic algorithm for solving large-scale SSCFLP instances that combines neighborhood-based heuristic procedures with the solution of two binary linear programming sub-problems through a general-purpose solver. The proposed algorithm starts from the optimal solution of the linear relaxation of the SSCFLP to reduce its size and identify promising potential locations for opening facilities. Computational experiments were performed on two benchmark sets of large instances. For one of them, the developed algorithm obtained optimal solutions for all instances. For the other set, it provided average relative deviations slightly lower than those of three relevant algorithms from the literature. These results allow us to conclude that the proposed algorithm generates good-quality solutions and is competitive in solving large-scale SSCFLP instances. [ABSTRACT FROM AUTHOR]

This paper studies the multi-trip vehicle routing problem with multiple time windows, which extends the multi-trip vehicle routing problem by deciding not only the sequence of customers that each vehicle serves but also the service time window of each customer. It also requires that the delivery service time is within the selected time windows and that the total demand of the customers served by the vehicle on each trip does not exceed the maximum carrying capacity. For solving the studied problem, we develop a mixed integer linear programming model with the objective of minimizing the total travel distance of vehicles and design a tailored iterative local search heuristic. Within the framework of the iterative local search, an improved Solomon greedy insertion algorithm suitable for multiple time windows and multi-trip scenarios is designed to generate the initial solution, and local search operators such as Or-opt and Relocate, as well as Random Exchange perturbation operations, are also developed. The experiment results demonstrate the effectiveness of the proposed model and algorithm and confirm that by providing customers with multiple time windows option, carriers can flexibly plan vehicle routes and select appropriate service time windows, thereby reducing the number of vehicles used and the total distance travelled and improve delivery success. [ABSTRACT FROM AUTHOR]

Copyright of Control Theory & Applications / Kongzhi Lilun Yu Yinyong is the property of Editorial Department of Control Theory & Applications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This paper presents a real-world auto-carrier loading and routing problem encountered by an automotive logistics company in China. The problem requires determining a set of routes for auto-carriers to transport vehicles from warehouses to dealers subject to several loading and routing constraints. Four objectives are considered, including the maximization of the number of loaded urgent vehicles and the number of loaded vehicles, and the minimization of the number of visits to dealers and travel distance. To solve such a multi-objective problem, we develop a two-stage iterated local search heuristic with the lexicographic optimization principle. Expressly, we assume that all vehicles are stored in one warehouse and solve the resulting problem as the first stage. Then the best-found solution is improved by considering the multiple warehouse constraints in the second stage. The effectiveness of the proposed heuristic is validated by comparing it with a commercial solver and the company’s approach. Moreover, computational results provide managerial insights for decision-makers when considering the objectives and their priorities in different scenarios. [ABSTRACT FROM AUTHOR]

Project schedule management, which involves generating and revising a schedule, is a critical decision-making process in uncertain environments. However, methods that can be used in project schedule management, particularly for large and complex projects, have so far been insufficiently studied. We believe that addressing this issue is essential to enable this process to be successfully adapted to practical projects. This study proposes a scheduling method that can generate a robust schedule insensitive to delays. The proposed method accurately evaluates schedule robustness using simulations, even for complex projects involving multiple decisions. As the simulation element of this process can be time-consuming, particularly for large-scale project instances, it is necessary to accelerate the evaluation process and ensure the search is performed efficiently. The proposed method reduces the computational time required for the evaluation process by employing parallel processing on a graphics processing unit (GPU) and utilizing simple simulations. To search for various solutions efficiently, the kick operation was developed that is specifically tailored to address the target problem. Numerical experiments confirmed the fundamental properties of the proposed method and demonstrated its effectiveness in comparison with previously employed methods in this area.

This paper addresses to the problem of flowshop scheduling under no-idle constraints with respect to the total completion time criterion. A branch and bound algorithm is proposed and a machine based lower bound is developed for the first time in the literature of this problem. Moreover, an Iterated Local Search (ILS) algorithm is used to provide the upper bounds of the problem. The computational results are promising, and the proposed algorithm appears able to find optimal solutions to problems with up to 20 jobs and 4 machines. [ABSTRACT FROM AUTHOR]

We address an inventory-routing problem that arises in a liquid oxygen-producing company. Decisions must be made for the efficient transport of the product from sources to destinations by means of a heterogeneous fleet of trucks. This combinatorial problem has been stated as a constrained minimization one, whose objective function is the quotient of the operating cost divided by the total amount of delivered product. The operating cost comes from the distances traveled, the drivers' salary, and the drivers' overnight accommodation. The constraints include time windows for drivers and destinations, inventory safety levels, lower bounds for the quantity of product delivered to destinations, and maximum driving times. To approximate the optimal solution of this challenging problem, we developed a heuristic algorithm that first finds a feasible solution, and then iteratively improves it by combining the Metropolis criterion with local search. Our results are competitive with the best proposals in the literature. [ABSTRACT FROM AUTHOR]

This paper addresses a real‐world multi‐period orienteering problem arising in a large Italian company that needs to patrol an area in order to provide security services to a set of customers. Each customer requires different services on a weekly basis. Some services are mandatory, while others are optional. It might be impossible to perform all optional services, and each of them is assigned a score when performed. The challenge is to determine a set of routes, one per day, that maximizes a weighted sum of the total collected score and total working time, while meeting several operational constraints, including hard time windows, maximum riding time, minimum number of services performed, and minimum time between two consecutive visits for the same service at the same customer. To solve the problem, we propose an iterated local search that invokes at each iteration an inner variable neighborhood descent procedure. Computational tests performed on a large number of real‐world instances prove that the developed algorithm is very efficient, and finds in a short time solutions that are consistently better than those produced by a mathematical model, and those in use at the company. [ABSTRACT FROM AUTHOR]

Copyright of GeSec: Revista de Gestao e Secretariado is the property of Sindicato das Secretarias e Secretarios do Estado de Sao Paulo (SINSESP) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This research studies the Field Technician Scheduling Problem with Experience-Dependent Service Times (FTSP–EDST), involving three main features: matching maintenance tasks with available technicians, sequencing the tasks, and considering the experience-dependent service times. Given a limited number of technicians, the objective is to maximize the collected profit for servicing tasks. This study formulates the problem as a mixed-integer linear programming model and proposes a Modified Iterated Local Search (MILS) to solve the benchmark problem instances of various sizes. A set of FTSP–EDST instances is generated based on existing publicly accessible data, and MILS is utilized to solve these newly generated instances. Computational results confirm the effectiveness of MILS in solving FTSP–EDST. [ABSTRACT FROM AUTHOR]

This paper addresses the double traveling salesman problem with partial last‐in‐first‐out (LIFO) loading constraints. A vehicle picks up items in the pickup area and loads them into its container, a horizontal stack. Once all the pickup operations are done, the vehicle can deliver the items to the delivery area because pickup‐and‐delivery areas are geographically separated. Additionally, the loading and unloading operations also follow a partial LIFO policy. The aim is to find a Hamiltonian cycle that satisfies the loading and unloading policy described above in order to minimize the distance traveled by the vehicle in both areas and the cost of the rearrangements that are performed across the routes. The state‐of‐the‐art algorithm finds good solutions to the existing instances. However, finding good‐quality solutions in large instances requires longer computing times. Therefore, we propose an iterated local search that considers two additional components to the classical components, the intensification and restart procedure. The first helps in a more thorough examination of promising neighborhoods, while the second helps in the complete diversity of the solutions in the hopes of escaping the local optima. Moreover, a new unique decoding approach for the solution representation is proposed to effectively generate feasible solutions. The proposed algorithm's performance is evaluated on an existing set of instances and in two new additional sets of larger instances that are proposed. The computational results obtained on the three test sets show that the proposed algorithm outperforms the state‐of‐the‐art algorithm in terms of computing time required to find good‐quality solutions. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we solve the well-known symbolic regression problem that has been intensively studied and has a wide range of applications. To solve it, we propose an efficient metaheuristic-based approach, called RILS-ROLS. RILS-ROLS is based on the following two elements: (i) iterated local search, which is the method backbone, mainly solving combinatorial and some continuous aspects of the problem; (ii) ordinary least squares method, which focuses on the continuous aspect of the search space—it efficiently determines the best—fitting coefficients of linear combinations within solution equations. In addition, we introduce a novel fitness function that combines important model quality measures: $$R^2$$ R 2 score, RMSE score, size of the model (or model complexity), and carefully designed local search, which allows systematic search in proximity to candidate solution. Experiments are conducted on the two well-known ground-truth benchmark sets from literature: Feynman and Strogatz. RILS-ROLS was compared to 14 other competitors from the literature. Our method outperformed all 14 competitors with respect to the symbolic solution rate under varying levels of noise. We observed the robustness of the method with respect to noise, as the symbolic solution rate decreases relatively slowly with increasing noise. Statistical analysis of the obtained experimental results confirmed that RILS-ROLS is a new state-of-the-art method for solving the problem of symbolic regression on datasets whose target variable is modelled as a closed-form equation with allowed operators. In addition to evaluation on known ground-truth datasets, we introduced a new randomly generated set of problem instances. The goal of this set of instances was to test the sensitivity of our method with respect to incremental equation sizes under different levels of noise. We have also proposed a parallelized extension of RILS-ROLS that has proven adequate in solving several very large instances with 1 million records and up to 15 input variables.

The success of a company requires the proper functioning and reliability of its systems with machines and equipment in good condition. For this, a good preventive maintenance plan is essential, which tends to become more complex as the number of equipment and the planning horizon increases. The present work aims to develop efficient meta-heuristic algorithms for the Long-Term Preventive Maintenance Scheduling Problem (PPOMPLP). The work begins with the development of a constructive and allocation heuristic, followed by the development of local search and meta-heuristic algorithms based on Greedy Randomized Adaptive Search Procedure (GRASP), Simulated Annealing (SA), and Iterated Local Search (ILS). The performance of the proposed algorithms was compared among themselves and with those of other studies in the literature. Small fictitious instances were used to calibrate and validate the meta-heuristic algorithms. After calibration, they were applied to solve larger and real instances. The experiments showed that the ILS was the best-performing algorithm, and its result for the real instance was 40.5% better than that presented in the literature. [ABSTRACT FROM AUTHOR]

In many major cities, metro lines constitute the backbone of urban public transport, providing an efficient and greener alternative to private mobility. An important feature that distinguishes metro lines from other public transport means, such as buses, is that metros are typically tightly resource constrained. The trains operating on a particular line are often specifically fitted for that line, making any capacity expansion extremely costly and time-consuming. Therefore, researchers and operators alike are seeking ways to make better use of existing resources. One possible way of doing so is by adapting timetables to forecasted demand while accounting for limited vehicle capacities. Thus, we consider a demand-driven nonperiodic timetabling problem for a two-directional metro line that minimizes the total passenger waiting time through the efficient scheduling of the available trains. Considering that passengers board trains using a well-mixed policy, we explicitly account for train capacities on a moment-to-moment basis. Last, we consider that trains are allowed to short turn. In this respect, we assume that trains must pass by a given station before short turning and are only allowed to idle after having short turned. We devise a polynomial time algorithm for assessing the total passenger waiting time generated by a given timetable and an effective lower bound that is evaluated in linear time. These are used in a variable neighborhood search algorithm, which is embedded in an iterated local search metaheuristic. Classical local search-based neighborhoods are not effective for our problem because they do not explicitly handle the vehicle scheduling decisions. To handle this challenge, we proposed three tailored neighborhoods. We validate our heuristic on the uncapacitated version of the problem. Considering a benchmark of 48 artificial instances with up to 20 stations, our heuristic achieved an average gap of 0.67% and found eight new best solutions. We also validated our heuristic on three sets of instances based on realistic lines from Milan, Madrid, and Beijing. Furthermore, we demonstrate the operational advantages of our optimized timetables in the capacitated version of the problem by comparing them with regular timetables and with exact solutions obtained for the uncapacitated case. Furthermore, we conduct a sensitivity analysis with respect to the capacity of the trains and investigate the impact of a priority boarding policy. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2022.0271. [ABSTRACT FROM AUTHOR]

The university course timetable problem (UCTP) is known to be NP-hard, with solution complexity growing exponentially with the problem size. This paper introduces an algorithm that effectively tackles UCTPs by employing a combination of exploration and exploitation strategies. The algorithm comprises two main components. Firstly, it utilizes a genetic algorithm (GA) to explore the search space and discover a solution within the global optimum region. Secondly, it enhances the solution by exploiting the region using an iterated local search (ILS) algorithm. The algorithm is tested on two common variants of UCTP: the post-enrollment-based course timetable problem (PE-CTP) and the curriculum-based course timetable problem (CB-CTP). The computational results demonstrate that the proposed algorithm yields competitive outcomes when compared empirically against other existing algorithms. Furthermore, a t-test comparison with state-of-the-art algorithms is conducted. The experimental findings also highlight that the hybrid approach effectively overcomes the limitation of local optima, which is encountered when solely employing GA in conjunction with local search. [ABSTRACT FROM AUTHOR]

The Traveling Salesman Problem (TSP) is a well known problem in operations research with various studies and applications. In this paper, we address a variant of the TSP in which the customers are divided into several priority groups and the order of servicing groups can be flexibly changed with a rule called the d-relaxed priority rule. The problem is called the Clustered Traveling Salesman Problem with Relaxed Priority Rule (CTSP-d). We propose two new Mixed Integer Linear Programming (MILP) models for the CTSP-d and a metaheuristic based on Iterated Local Search (ILS) with operators designed for or adapted to the problem. The experimental results obtained on the benchmark instances show that two new models performs better than previous ones, and ILS also proves its performance with 13 new best known solutions found and significant stability compared to existing metaheuristics. [ABSTRACT FROM AUTHOR]

The traveling salesman problem (TSP) is a very popular combinatorics problem. This problem has been widely applied to various real problems. The TSP problem has been classified as a Non-deterministic Polynomial Hard (NP-Hard), so a non-deterministic algorithm is needed to solve this problem. However, a non-deterministic algorithm can only produce a fairly good solution but does not guarantee an optimal solution. Therefore, there are still opportunities to develop new algorithms with better optimization results. This research develops a new algorithm by hybridizing three local search algorithms, namely, iterated local search (ILS) with simulated annealing (SA) and hill climbing (HC), to get a better optimization result. This algorithm aimed to solve TSP problems in the transportation sector, using a case study from the Traveling Salesman Challenge 2.0 (TSC 2.0). The test results show that the developed algorithm can optimize better by 15.7% on average and 11.4% based on the best results compared to previous studies using the Tabu-SA algorithm. [ABSTRACT FROM AUTHOR]

Multi-compartment vehicle routing problem (MCVRP) is an extension of the classical capacitated vehicle routing problem where products with different characteristics are transported together in one vehicle with multiple compartments. This paper deals with this problem, whose objective is to minimize the total travel distance while satisfying the capacity and maximum route length constraints. We proposed a hybrid iterated local search metaheuristic (HILS) algorithm to solve it. In the framework of iterated local search, the current solution was improved iteratively by five neighborhood operators. For every obtained neighborhood solution after the local search procedure, a large neighborhood search-based perturbation method was executed to explore larger solution space and get a better neighborhood solution to take part in the next iteration. In addition, the worse solutions found by the algorithm were accepted by the nondeterministic simulated annealing-based acceptance rule to keep the diversification of solutions. Computation experiments were conducted on 28 benchmark instances and the experimental results demonstrate that our presented algorithm finds 17 new best solutions, which significantly outperforms the existing state-of-the-art MCVRP methods. [ABSTRACT FROM AUTHOR]

In this paper, we solve the well-known symbolic regression problem that has been intensively studied and has a wide range of applications. To solve it, we propose an efficient metaheuristic-based approach, called RILS-ROLS. RILS-ROLS is based on the following two elements: (i) iterated local search, which is the method backbone, mainly solving combinatorial and some continuous aspects of the problem; (ii) ordinary least squares method, which focuses on the continuous aspect of the search space—it efficiently determines the best—fitting coefficients of linear combinations within solution equations. In addition, we introduce a novel fitness function that combines important model quality measures: R 2 score, RMSE score, size of the model (or model complexity), and carefully designed local search, which allows systematic search in proximity to candidate solution. Experiments are conducted on the two well-known ground-truth benchmark sets from literature: Feynman and Strogatz. RILS-ROLS was compared to 14 other competitors from the literature. Our method outperformed all 14 competitors with respect to the symbolic solution rate under varying levels of noise. We observed the robustness of the method with respect to noise, as the symbolic solution rate decreases relatively slowly with increasing noise. Statistical analysis of the obtained experimental results confirmed that RILS-ROLS is a new state-of-the-art method for solving the problem of symbolic regression on datasets whose target variable is modelled as a closed-form equation with allowed operators. In addition to evaluation on known ground-truth datasets, we introduced a new randomly generated set of problem instances. The goal of this set of instances was to test the sensitivity of our method with respect to incremental equation sizes under different levels of noise. We have also proposed a parallelized extension of RILS-ROLS that has proven adequate in solving several very large instances with 1 million records and up to 15 input variables. [ABSTRACT FROM AUTHOR]