Test Instances Research Articles

Optimal Task Allocation and Sequencing for Flight Test Based on a Memetic Algorithm With Lexicographic Optimisation

ABSTRACTThe flight test plays an important role in the development of an aircraft. Currently, with the increasing complexity and higher validation requirements for aircraft, there is a crucial need to generate high‐quality flight test task schedules in an efficient way. This paper proposes a flight test task scheduling problem (FTTSP), which involves assigning suitable aircraft and executing the flight test tasks in a given order. Generally, the flight test duration (FTD) is the primary optimisation objective for the flight test task schedule, as it has a direct impact on aircraft development costs and the time to enter the market. In this study, the FTTSP not only considers FTD but also takes into account task transfer consumption (TTC). A mixed‐integer linear programming mathematical model is first formulated to describe the FTTSP characteristics with the optimisation of the FTD and the TTC in a sequential manner. Then, a memetic algorithm with lexicographic optimisation (MALO) is proposed, which can efficiently obtain a high‐quality solution and ensure that the most critical metric can be fully optimised. In MALO, a two‐vector encoding and a task logic relationship repair mechanism based on the binary tree are established. An idle time insertion decoding method is designed to improve the aircraft utilisation rate. In addition to the selection, crossover and mutation operators, a local search operator is designed to enhance the solution quality. Finally, the full‐scale test instances are generated for the FTTSP to evaluate the algorithm's performance. The numerical results demonstrate the effectiveness and competitiveness of the MALO in generating a high‐quality schedule for flight test tasks.

A novel multi-objective artificial bee colony algorithm for solving the two-echelon load-dependent location-routing problem with pick-up and delivery

This study considers a two-echelon load-dependent location routing problem with pick-up and delivery (2E-LDLRPPD). As a variant of the two-echelon vehicle routing problem with pick-up and delivery (2E-VRPPD), the 2E-LDLRPPD includes additional variants such as two-echelon location-routing problem (2E-LRP) and load-dependent vehicle routing problem (LDVRP). However, much of the existing research work has traditionally focused on a single objective, predominantly aimed to minimize costs. In our case, we build a multi-objective model that concurrently minimizes costs, carbon emissions, and the number of vehicles used. Heuristic algorithms are commonly used to solve complex location-routing problems. Therefore, we propose a hybrid heuristic algorithm named the improved elite-guided multi-objective artificial bee colony algorithm with variable neighborhood search (IEMOABC-VNS). Base on elite-guided multi-objective artificial bee colony algorithm (EMOABC), a two-archive elite-guide strategy is deployed to strike a balance between diversity and convergence. The efficacy of the IEMOABC-VNS is compared experimentally with four other hybrid heuristic algorithms on test instances and a real-world case. Computational results demonstrate that the IEMOABC-VNS outperforms the competing algorithms in solving 2E-LDLRPPD, and obtains a high-quality Pareto front in a relatively short time. Especially, the algorithm exhibits significant performance enhancements when applied to large-scale instances.

Optimal operation strategies for freight transport with electric vehicles considering wireless charging lanes

Unlike traditional recharging at fixed stations, wireless charging lanes equipped on the road network enable electric vehicles (EVs) to recharge while in motion, reducing the time required for EV recharging. Indeed, the availability of wireless charging lanes allows the urban freight transport service operators to consider adjusting EVs routing plan to do EV recharging in transit so that the total operational costs can be further reduced. Envisioning that wireless charging lanes would be practically constructed on the road network, this study aims to investigate how the optimal EV routing plan can be determined in a pickup and delivery problem in the future electrified mobility system. The problem is formulated into a mixed-integer programming model, and we propose a branch-and-price algorithm to solve it. Additionally, a large neighborhood search heuristic is incorporated within the column generation framework to effectively tackle the sub-problems. Computational experiments on test instances highlight the effectiveness of our proposed algorithm and demonstrate the potential of wireless charging lanes to reduce total operational costs.

Towards Dataset-scale and Feature-oriented Evaluation of Text Summarization in Large Language Model Prompts.

Recent advancements in Large Language Models (LLMs) and Prompt Engineering have made chatbot customization more accessible, significantly reducing barriers to tasks that previously required programming skills. However, prompt evaluation, especially at the dataset scale, remains complex due to the need to assess prompts across thousands of test instances within a dataset. Our study, based on a comprehensive literature review and pilot study, summarized five critical challenges in prompt evaluation. In response, we introduce a feature-oriented workflow for systematic prompt evaluation. In the context of text summarization, our workflow advocates evaluation with summary characteristics (feature metrics) such as complexity, formality, or naturalness, instead of using traditional quality metrics like ROUGE. This design choice enables a more user-friendly evaluation of prompts, as it guides users in sorting through the ambiguity inherent in natural language. To support this workflow, we introduce Awesum, a visual analytics system that facilitates identifying optimal prompt refinements for text summarization through interactive visualizations, featuring a novel Prompt Comparator design that employs a BubbleSet-inspired design enhanced by dimensionality reduction techniques. We evaluate the effectiveness and general applicability of the system with practitioners from various domains and found that (1) our design helps overcome the learning curve for non-technical people to conduct a systematic evaluation of summarization prompts, and (2) our feature-oriented workflow has the potential to generalize to other NLG and image-generation tasks. For future works, we advocate moving towards feature-oriented evaluation of LLM prompts and discuss unsolved challenges in terms of human-agent interaction.

Bayesian Learning Based Elitist Nondominated Sorting Algorithm for a Kind of Multi-objective Integrated Production Scheduling and Transportation Problem

This paper focuses on a kind of multi-objective integrated production scheduling and transportation problem (MIPSTP), which is encountered in many real-life industrial processes. MIPSTP contains a production stage for processing products and a transportation stage for transporting products. In MIPSTP, we consider a dedicated integration of distributed scheduling and transportation regarding the two stages, arising from a practical project for a cooperative iron and steel company. The objective of MIPSTP is to minimize the total completion time of each factory and total transportation cost. To solve the problem, a Bayesian learning-based elitist nondominated sorting algorithm (BLENSA) is proposed. The primary highlights of this work are two-fold: 1) new solution structures of MIPSTP and 2) novel search model of BLENSA. For the solution structures, we propose for the first time the mathematical description of MIPSTP, accounting for several features in applications and so is different from conventional scheduling problems. For the search model of BLENSA, we propose a Bayesian learning-based probability model to learn valuable knowledge about nondominated solutions. Thereby, BLENSA combines the Bayesian learning-based probability model to produce a candidate population (CP) and the nondominated sorting method to generate a main population (MP). Next, we propose a greedy competition strategy to construct MP from CP and MP for the next generation. Results of experiments on 57 test instances and a real-life case study demonstrate the effectiveness and practical values of BLENSA.

Division-selection transfer learning for prediction based dynamic multi-objective optimization

Dynamic multi-objective optimization problems (DMOPs) are challenging as they require capturing the Pareto optimal front (POF) and Pareto optimal set (POS) during the optimization process. In recent years, transfer learning (TL) has emerged the empirical knowledge and is an effective approach to solve DMOPs. However, negative transfer can occur when the transfer method is not suitable for the transfer task. It may deviate the search path and seriously reduce the efficiency, so how to reduce the occurrence of negative transfer to save the running time of dynamic multi-objective evolutionary algorithm (DMOEA) is an important issue to be addressed. A division-selection transfer learning evolutionary algorithm for dynamic multi-objective optimization (DST-DMOEA) is designed towards this aim. Specifically, individuals with high Spearman correlation are relatively stable in different environments, selecting them to train the Support Vector Regression (SVR) model ensures a more accurate capture of solution features, predicting the objective values of historical solutions based on the model, and thus divide historical solutions into elite and non-elite solutions. Subsequently, for the elite solutions, individual TL that incorporates local information for optimization and transfer is used, while the non-elite solutions are handled with manifold TL method to obtain the overall data distribution and understand the internal structure. Then, merge the predicted individuals generated by two parts of TL will constitute as the initial population in the optimization process. Compared with other algorithms, the initial solution of DST-DMOEA is closer to the real POF, effectively reducing negative transfer. In addition, in 51 test instances, DST-DMOEA has shown superior performance in over 30 instances.

Evaluating time-cost trade-off in a resource constraint GERT-type project scheduling problem considering rework

In practice, rework is a major threat to projects’ performance and increases the risk of not delivering the outcomes on time and budget. The classical project scheduling methods are incapable of considering the rework and stochastic structure for projects in the planning phase. The current study models the stochastic projects through a GERT network in which the occurrence of rework is allowed. The aim is to find the optimal execution mode for each activity so that the project is completed within the desired time and the resource constraint is satisfied. We develop a multi-mode time–cost trade-off resource constraint project scheduling problem under uncertainty to find a here-and-now solution to minimise project completion time. The problem is solved using a combination of flow-graph theory and a branch and bound method for the developed integer programming model. As the project structure is stochastic, a distinct set of all possible realised paths is considered. To the best of our knowledge, there is no similar test instance in the literature, hence, the solution methodology is implemented on a randomly generated numerical example. The reported computational results exhibit the sensitivity of objective function to the changes in available resource amount.

Task offloading for Industrial Internet of Things based on evolutionary multiobjective multitasking optimization

In mobile edge computing(MEC) scenarios, offloading computation-intensive and delay-critical tasks to nearby mobile edge servers with abundant computing resources can reduce the time delay(TD) of tasks and energy consumption(EC) of terminal devices, thus improving application performance and user experience. In order to reduce the TD of tasks and the EC of factory terminal devices(FTDs), MEC is introduced into the Industrial Internet of Things(IIoT). Because TD and EC are conflicting objectives, most existing research models the task offloading problem(TOP) as a multiobjective optimization problem to optimize both TD and EC. However, as the number of FTDs and tasks in the MEC network increases, the scale and computational cost of the TOP increases, and it is very challenging to obtain the optimal task offloading scheme through multiobjective optimization. Multitasking optimization can make use of knowledge transfer between related tasks to promote the solving efficiency of each task. Based on this, we model the TOP in IIoT as a multiobjective TOP (MTOP). A cheap task that is highly similar to MTOP is constructed, and a novel evolutionary multiobjective multitasking framework is developed, which uses the positive knowledge transfer between tasks to optimize TD and EC in the network. Then, an effective evolutionary multiobjective multitasking algorithm is proposed, which includes multi-functional knowledge transfer strategy(MFKT) and adaptive cheap task update strategy(ACTU). MFKT uses decision variables with different functions in the cheap tasks for positive knowledge transfer, so as to improve the performance of MTOP. ACTU dynamically updates the cheap task to maintain knowledge transfer between tasks. In different test instances, the proposed algorithm is compared with the existing multiobjective and multitasking algorithms. The experimental results show that the proposed algorithm is more competitive in terms of TD and EC, it can guarantee the service quality of the IIoT system and has strong robustness and scalability.

An enhanced guided stochastic search with repair deceleration mechanism for very high-dimensional optimization problems of steel double-layer grids

Finding reasonably good solutions using a fewer number of objective function evaluations has long been recognized as a good attribute of an optimization algorithm. This becomes more important, especially when dealing with very high-dimensional optimization problems, since contemporary algorithms often need a high number of iterations to converge. Furthermore, the excessive computational effort required to handle the large number of design variables involved in the optimization of large-scale steel double-layer grids with complex configurations is perceived as the main challenge for contemporary structural optimization techniques. This paper aims to enhance the convergence properties of the standard guided stochastic search (GSS) algorithm to handle computationally expensive and very high-dimensional optimization problems of steel double-layer grids. To this end, a repair deceleration mechanism (RDM) is proposed, and its efficiency is evaluated through challenging test examples of steel double-layer grids. First, parameter tuning based on rigorous analyses of two preliminary test instances is performed. Next, the usefulness of the proposed RDM is further investigated through two very high-dimensional instances of steel double-layer grids, namely a 21,212-member free-form double-layer grid, and a 25,514-member double-layer multi-dome, with 21,212 and 25,514 design variables, respectively. The obtained numerical results indicate that the proposed RDM can significantly enhance the convergence rate of the GSS algorithm, rendering it an efficient tool to handle very high-dimensional sizing optimization problems.

A Reduced Modeling Approach for Making Predictions with Incomplete Data Having Blockwise Missing Patterns

Incomplete data with blockwise missing patterns are commonly encountered in analytics, and solutions typically entail listwise deletion or imputation. However, as the proportion of missing values in input features increases, listwise or columnwise deletion leads to information loss, whereas imputation diminishes the integrity of the training data set. We present the blockwise reduced modeling (BRM) method for analyzing blockwise missing patterns, which adapts and improves on the notion of reduced modeling proposed by Friedman, Kohavi, and Yun in 1996 as lazy decision trees. In contrast to the original idea of reduced modeling of delaying model induction until a prediction is required, our method is significantly faster because it exploits the blockwise missing patterns to pretrain ensemble models that require minimum imputation of data. Models are pretrained over the overlapping subsets of an incomplete data set that contain only populated values. During prediction, each test instance is mapped to one of these models based on its feature-missing pattern. BRM can be applied to any supervised learning model for tabular data. We benchmark the predictive performance of BRM using simulations of blockwise missing patterns on three complete data sets from public repositories. Thereafter, we evaluate its utility on three data sets with actual blockwise missing patterns. We demonstrate that BRM is superior to most existing benchmarks in terms of predictive performance for linear and nonlinear models. It also scales well and is more reliable than existing benchmarks for making predictions with blockwise missing pattern data. Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijds.2022.9016 .

An Improved Combinatorial Benders Decomposition Algorithm for the Human-Robot Collaborative Assembly Line Balancing Problem

As an emerging technology, human-robot collaboration (HRC) has been implemented to enhance the performance of assembly lines and improve the safety of human workers. By integrating the advantages of human workers and collaborative robots (cobots), HRC enables production systems to process tasks consecutively, concurrently, or collaboratively. However, the introduction of cobots also makes the corresponding human-robot collaborative assembly line balancing problem more complex and difficult to solve. To solve this problem, we first propose an enhanced mixed integer program (EMIP) with various enhancement techniques and tighter bounds, and then, we develop an improved combinatorial Benders decomposition algorithm (Algorithm ICBD) with new local search strategies, Benders cuts, and acceleration procedures. To verify the effectiveness of our proposed model and algorithms, we conduct extensive computational experiments, and the results show that our proposed EMIP model is significantly better than the existing mixed integer program model; the percentages of instances that can obtain feasible and optimal solutions are increased from 82.42% to 100% and from 29.17% to 43.5%, respectively, whereas the average gap is decreased from 19.81% to 5.64%. In addition, our proposed Algorithm ICBD can get 100% of feasible solutions and 65.92% of optimal solutions for all of the test instances, and the average gap is only 1.49%. Moreover, compared with existing Benders decomposition methods for this problem, our approach yields comparatively better solutions in notably shorter average computational time when run in the same computational environment. History: Accepted by Andrea Lodi, Area Editor for Design & Analysis of Algorithms–Discrete. Funding: This research was supported by the National Natural Science Foundation Council of China [Grants 72401214, 92167206, 7221101377, 72471169, and 72231005], the Ministry of Education of China [Grant 24YJC630078], and Computation and Analytics of Complex Management Systems (Tianjin University). This research was also supported by the Tianjin Natural Science Foundation Project [Grant 23JCQNJC01900] and the Tianjin Philosophy and Social Science Planning Project [Grant TJGL21-016]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.0279 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2023.0279 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Multi-objective grey wolf optimizer based on reinforcement learning for distributed hybrid flowshop scheduling towards mass personalized manufacturing

As an emerging production paradigm, mass personalized manufacturing (MPM) realizes personalized product customization under the premise of ensuring large-scale production. In this paradigm, the rapid switching of the type and quantity of manufacturing tasks increases the difficulty of scheduling. Hence, this paper proposes the distributed hybrid flowshop scheduling problem with an order modularization and tasks assigning method (DHFSP-OMTA), where heterogeneous customer orders are decomposed into standard and personalized production tasks and assigned to different factories. Meantime, towards MPM, a novel mixed integer linear programming model is established to minimize the makespan and total energy consumption simultaneously. Considering the high complexity of DHFSP-OMTA, a multi-objective grey wolf optimizer based on reinforcement learning (MOGWO-RL) is designed. This paper contains the following three improvements. Firstly, the variable tasks splitting method combines two initial heuristic-rule to produce a high-quality population. Secondly, a variable neighborhood search based on reinforcement learning is designed to improve the search quality and jump out of the local optimum. Thirdly, an efficient merging batches method is presented to save transportation energy consumption. The advantages of the proposed algorithm are verified on 18 modified test instances based on the Taillard benchmark with the MPM feature. The results show that MOGWO-RL has the best effectiveness and stability of all comparison algorithms. Therefore, it can be used as a novel method to solve MPM’s scheduling problem.

An improved non-dominated sorting genetic algorithm II for distributed heterogeneous hybrid flow-shop scheduling with blocking constraints

Distributed manufacturing is a new trend to accommodate the economic globalization, which means multiple geographically-distributed factories can collaborate to meet urgent delivery requirements. However, such factories may vary due to layout adjustments and equipment aging, thus the production efficiency greatly depends on the allocation of orders. This scenario is frequently found in energy-intensive process industries, e.g., chemical and pharmaceutical and industries, where the lack of buffers usually results in extra non-blocking constraints and makes the production scheduling even harder. Therefore, this paper investigates a distributed heterogeneous hybrid blocking flow-shop scheduling problem (DHHBFSP) for minimizing the makespan and total energy consumption simultaneously, and proposes an improved non-dominated sorting genetic algorithm II (INSGA-II) to address the problem. First, two heuristic algorithms, i.e., bi-objective considered heuristic (BCH) and similarity heuristic (SH), are developed for the population initialization. Then, to speed-up the local search, a comparison method for non-dominated solutions is proposed to reserve more solutions that are likely to be further improved. Afterwards, a probabilistic model is developed to eliminate unnecessary operations during local search processes. Finally, the proposed INSGA-II is tested on benchmark instances and a real-world case for the validation. Numerical experiments suggest that the SH can generates high-quality solutions within a very short period of time, and the BCH has significantly improved average IGD and HV values for the initial population. Besides, the probabilistic model saves considerable computational time for the local search without compromising the solution quality. With the help of these strategies, the proposed INSGA-II improves average IGD and HV values by 68 % and 57 % for the basic NSGA-II respectively, and obtains better Pareto fronts compared to existing multi-objective algorithms on the majority of test instances. Moreover, the industrial case study shows that the proposed INSGA-II is capable of providing solid scheduling plans for a pharmaceutical enterprise with large-scale orders.

Correcting the distribution of batch normalization signals for Trojan mitigation

Backdoor (Trojan) attacks represent a significant adversarial threat to deep neural networks (DNNs). In such attacks, the presence of an attacker’s backdoor trigger causes a test instance to be misclassified into the attacker’s chosen target class. Post-training mitigation methods aim to rectify these misclassifications, ensuring that poisoned models correctly classify backdoor-triggered samples. These methods require the defender to have access to a small, clean dataset and the potentially compromised DNN. However, most defenses rely on parameter fine-tuning, making their effectiveness dependent on the dataset size available to the defender. To overcome the limitations of existing approaches, we propose a method that rectifies misclassifications by correcting the altered distribution of internal layer activations of backdoor-triggered instances. Distribution alterations are corrected by applying simple transformations to internal activations. Notably, our method does not modify any trainable parameters of the DNN, yet it achieves generally good mitigation performance against various backdoor attacks and benchmarks. Consequently, our approach demonstrates robustness even with a limited amount of clean data, making it highly practical for real-world applications. The effectiveness of our approach is validated through both theoretical analysis and extensive experimentation. The appendix is provided as an electronic component and can be accessed via the link in the footnote.22https://arxiv.org/pdf/2308.09850. The source codes can be found in the link33https://github.com/lixi1994/BNA. at the footnote.

Real-Time Coordination and Communication Process for the Issuance of Emergency Ambulance Alerts

The lack of attention to emergency alerts in road accidents in a timely manner may be due to the lack of management of resources or the lack of communication between hospital and patrol departments. The purpose of this research is to reduce the management time to provide attention to requests on highways. An alert and resource management process is proposed. This process will allow sending the optimized information in the shortest possible time. The experimentation was worked with a set of test instances and simulated in the laboratory using mechatronic devices to show the communication, coordination and attention in real time. The results obtained were acceptable, it was possible to identify the alert to be included in the algorithm which must comply with the restrictions and subsequently calculate the optimal route to be followed by the mechatronic device to reach the request node for the attention of the distress alert.

Earthquake magnitude prediction in Indonesia using a supervised method based on cloud radon data

In the challenging realm of earthquake prediction, the reliability of forecasting systems has remained a persistent obstacle. This study focuses on earthquake magnitude prediction in Indonesia, leveraging supervised machine learning techniques and cloud radon data. We present an analysis of the tele-monitoring system, data collection methods, and the application of regression-based machine learning algorithms. Utilizing a comprehensive dataset spanning 30 training instances and 105 test instances, the study evaluates multiple metrics to ascertain the efficacy of the prediction models. Our findings reveal that the linear regression approach yields the best earthquake magnitude prediction method, with the lowest values across multiple evaluation metrics: standard deviation 0.40, mean absolute error (MAE) 0.30, mean absolute percentage error (MAPE) 6%, root mean square error (RMSE) 0.52, mean squared error (MSE) 0.28, symmetric mean absolute percentage error (SMAPE) 0.06, and conformal normalized mean absolute percentage error (cnSMAPE) 0.97. Additionally, we discuss the implications of the research results and the potential applications in enhancing existing earthquake prediction methodologies.

A speed-up procedure and new heuristics for the classical job shop scheduling problem: A computational evaluation

The speed-up procedure proposed for the permutation flowshop scheduling problem with makespan minimisation (commonly denoted as Taillard’s acceleration) remains, after 30 years, one of the most important and relevant studies in the scheduling literature. Since its proposal, this procedure has been included in countless approximate optimisation algorithms, and its use is mandatory for several scheduling problems. Unfortunately, despite the importance of such a procedure in solving scheduling problems, we are not aware of any related speed-up procedure proposed for the classical job-shop scheduling problem. First, this study aims to fill this gap by proposing a novel speed-up procedure for the job-shop scheduling problem with makespan minimisation, capable of reducing the complexity of insertion-based procedures n times. Second, to test its performance, the procedure is embedded in a critical-path-based local search method. Furthermore, we thirdly propose five constructive and composite heuristics to obtain high-quality solutions in short time intervals. The composite heuristics apply the previous procedure to reduce their computational efforts. Finally, to complete the study, we conduct an extensive computational evaluation on 243 test instances from eight distinct benchmarks. In this evaluation, 30 heuristics are re-implemented and compared under the same computer conditions. The results indicate the superiority of the proposed approaches compared to the competitive algorithms for the problem under study.

A Multi‐Trip Vehicle Routing Problem With Release Dates and Interrelated Periods

ABSTRACTThis article proposes a new multi‐trip vehicle routing problem variant, originally motivated by a practical application, namely a car components distribution to repair centers (warehouses). The problem is named the multi‐trip vehicle routing with release dates and interrelated periods (MTVRP‐RDIP). The routes may start at different pre‐defined periods, called departure periods, and may have different durations. Thus, two routes that start at different periods may be active at the same period, leading to the so‐called interrelated periods. Moreover, a route may only start at a period if a vehicle is available, and the availability of the vehicle depends not only on the existing vehicles but also on the active routes at that time period. The clients are classified according to their importance and represent warehouses that require car components that are made available throughout the time horizon, thus leading to the consideration of release dates. Delays in satisfying client orders and not serving orders in the time horizon result in penalty costs that must be minimized. The objective to minimize also includes routing costs and vehicle utilization costs. Two metrics are defined to compute an order's delay, each leading to a different mixed integer linear model. Computational results showed that one model clearly outperforms the other and even this one is only suited to address the smallest instances. A matheuristic based on a rolling‐horizon process that iteratively solves the better‐performing model with fewer periods was designed to tackle the largest instances. The matheuristic can provide feasible solutions for all test instances in an efficient manner that are, on average, better than the ones provided by the model in most of the compared key performance indicators.

Statistical analyses of solution methods for the multiple-choice knapsack problem with setups: Implications for OR practitioners

An interesting extension of the classic Knapsack Problem (KP) is the Multiple-Choice Knapsack Problem with Setups (MCKS) which is focused on solving practical applications that involve both multiple periods and setups. Sophisticated solution methods for the MCKS that are presented in the operations research (OR) literature are not readily available for use by OR practitioners. Using MCKS test instances that appear in the literature, we demonstrate that the general-purpose integer programming software Gurobi sometimes used in an iterative manner can efficiently solve these MCKS instances using all default parameter values on a standard PC. It is shown both empirically and statistically that these Gurobi solutions are competitive with solution approaches from the literature. Hence, our approach using Gurobi is both easy for the OR practitioner to use and gives results competitive with the best specialized MCKS solution methods in the literature without the need to generate algorithm-specific code. Furthermore, this paper presents significant concerns regarding the solutions stated in the literature by the approximate solution method that reports the best results on 120 MCKS test instances. Specifically, 26% of this method’s solutions violate Gurobi upper bounds and an additional 33% of its solutions, on average, exceed the known guaranteed optimums by a value of 12,510.

Open shop scheduling with group and transportation operations by learning-driven hyper-heuristic algorithms

Open shop scheduling problems (OSSPs) are complex scheduling problems, which have been extensively studied in the literature. Group and transportation activities are two important aspects of OSSPs that still need attention. This work considers an OSSP with group and transportation operations to minimize maximum completion time by solving three key sub-problems: job assignment among groups, job sequence in groups and group sequence on machines. Firstly, an integer programming model is formulized to define the problem. Secondly, a learning-driven hyper-heuristic algorithm is developed by incorporating a Q-learning method and four meta-heuristics, i.e., genetic algorithm, artificial bee colony optimization, variable neighborhood search method and Jaya algorithm. The Q-learning method is devised to select the most promising meta-heuristic for performing at each iteration. Three neighborhood structures are designed by integrating critical machines and critical paths. Finally, the developed model is verified by an exact solver CPLEX, and the comparison results exhibit that CPLEX is effective for instances with ten jobs. For the instances with more than ten jobs, the developed algorithm wins CPLEX in terms of computation accuracy and efficiency, signifying its excellent performance in finding better solutions. Furthermore, four meta-heuristics mentioned above and three state-of-the-art meta-heuristics are employed for comparisons in solving a set of benchmark test instances. The results confirm that the formulated model and algorithm have stronger competitiveness in handling the considered problems.