Multiple Heuristics Research Articles

Quantifying heuristic evaluation

Since Nielsen presented his work on heuristic evaluation in the 90s, it has been seen how this inspection method has been used recurrently not only in academia but also in the industry. The benefits of their application have been studied widely and multiple heuristics for multiple purposes have been designed. Despite it, quantifying their qualitative nature has been challenging. This research presents a novel approach to provide a framework for quantifying an heuristic evaluation based on the calculus of what we called Usability Percentage. We have used an initial study that started in 2018 with 7 different evaluators and we presented a new methodology to calculate the aforementioned percentage and also provide multiple information that might be added into a final report. We have used the platform RUXAILAB that allows us to create heuristic evaluations to prepare and execute the test. Our findings are that by comparing both studies the new methodology presents a narrowed data distribution through the different evaluators and allows us to identify in a more precise and concise way which heuristics might fail in an evaluation test.

A GRA-based approach for optimal selection of SALBP-1 assembly line balancing heuristics

Assembly line balancing is a compulsory approach which enables mass production of similar parts with proper efficacy. Multiple heuristics are proposed in literature to solve single model assembly line balancing problems (SALBP-1). Previous work has focused on developing and evaluating assembly line balancing heuristics and apply them to different fields. Other work investigates the factors affecting the performance of particular assembly line balancing heuristic. However, there is still a need for a guideline for selecting the most suitable heuristic for a particular application with insights into multiple performance measures. This research presents a grey relational analysis (GRA)-based approach to aid process designers in selecting the heuristic generating that yields the best line balancing. Specifically, combinations of factors of cycle time, order strength, number of tasks, and time variability are considered as operational setups of the line balancing process. The solution of each setup is solved by each heuristic, and the performance of the solution is evaluated. Then, GRA is applied to rank the heuristics on basis of the resulting grey rational grades. The proposed approach provides useful insights into the combinations suitable for a particular heuristic. In this study, time variability and cycle time are shown to have the greatest impact on the overall performance of the considered heuristics.

A toolbox to evaluate the trustworthiness of published findings

During the past few years, researchers have criticized their professions for providing an entry point for false-positive results arising from publication bias and questionable research practices such as p-hacking (i.e., selectively reporting analyses that yield a p-value below 5 %). Researchers are advocating replication studies and the implementation of open-science practices, like preregistration, in order to identify trustworthy effects. Nevertheless, because such consumer research developments are still emerging, most prior research findings have not been replicated, leaving researchers in the dark as to whether a line of research or a particular effect is trustworthy. We tackle this problem by providing a toolbox containing multiple heuristics to identify data patterns that might, from the information provided in published articles, indicate publication bias and p-hacking. Our toolbox is an easy-to-use instrument with which to initially assess a given set of findings.

The INCOSE Systems Engineering Heuristics: What Are They Telling Us About the Discipline?

AbstractThe INCOSE Fellows have been leading an effort to develop heuristics that will help systems engineers pass on their knowledge of the best practices within the discipline.We performed bibliometric analysis on the text corpus of 164 heuristics and of the sources cited in them to gain insights into what the heuristics data is telling us about the discipline of systems engineering.We performed co‐occurrence analysis using VOSviewer, which is a software application that identifies pairs of terms and connects them. This analysis found terms that occur together across multiple heuristics by applying a natural language processing algorithm that identified the terms and formed clusters of terms that are strongly related to each other. To gain further insight into the cooccurrence results, the INCOSE Fellows participated in a contest to name the clusters and create narratives that provide a rationale for the cluster names and why it makes sense for the terms to belong to the clusters they were assigned to. Finally, a ballot was created to send out to a broader collection of INCOSE members to vote on the best name for each cluster among the names provided by the Fellows. The three‐step process of bibliometric analysis, cluster naming by the Fellows, and voting by the members identified emergent disciplines essential to the practice of systems engineering.We also performed bibliographic coupling analysis and co‐citation analysis using VOSviewer. A bibliographic coupling link is a link between two heuristics that both cite the same source. A co‐citation link is a link between two sources that are both cited by the same heuristic. The coupling analysis identified clusters among the heuristics that revealed communities of practice within systems engineering, and the co‐citation analysis identified clusters among the cited references to reveal communities of theory in systems engineering.

Tourist scams: cues and processes in decision-making

ABSTRACT Tourist scams have long been a challenge for tourist security and destination management. Understanding the tourist decision-making process can improve scam prevention. Through a quasi-experiment, 609 participants made, and documented reasons for, decisions after viewing animated scam videos. Thematic analysis was conducted to explore the external (operational) cues and internal (respondent-related) references, and revealed two categories of cues, namely scam-covering and scam-revealing cues. Three internal references—emotions, knowledge and needs—exerted a crucial impact. Multiple heuristics strategies were identified, and a model was made to illustrate the decision-making process. Theoretic and pragmatic implications are discussed for communicating scam detection.

The applications of multiple route optimization heuristics and meta-heuristic algorithms to solid waste transportation: A case study in Turkey

The dense population and high solid waste production make it challenging to determine the optimal routes for garbage collection trucks in metropolitan cities. Moreover, incorrectly planned routes cause waste of fuel, time, and other resources in the municipal garbage collection process. Vehicle Routing Problem (VRP) is a combinatorial optimization problem that aims to find optimal routes for a certain number of vehicles that start from a depot, visit all customers, and return to the depot. This study aims to find the optimal routes that satisfy all the needs by considering the waste collection process of the Istanbul Umraniye district. Data on the currently used routes, daily tonnages, number of truck trips, and the number and locations of containers in the Umraniye district were obtained from the Umraniye municipality. We solve the route optimization problem of the Umraniye municipality with multiple heuristics and meta-heuristic algorithms like the nearest neighbor, savings, Christofides, simulated annealing, greedy descent, guided local search, and Tabu Search algorithms. We show the examined heuristics and meta-heuristics can produce fast and feasible solutions to large-size real-life problems.

Transportation Labor Cost Reduction via Vehicle Platooning: Alternative Models and Solution Methods

The advent of automated and semiautomated driving technologies is prompting a reevaluation of the benefits of truck platooning. Whereas the effect of truck platooning on fuel costs is relatively well studied, automated driving enables further cost reductions in the form of labor savings that has, heretofore, been sparingly investigated. The degree to which such savings can be realized depends upon myriad factors. Foremost among them are driver payment policies and platoon formation procedures. Automated driving technology reduces the demands on a platoon’s follower vehicles, which, in turn, justifies modified driver compensation structures. However, reductions in labor expenditures are often limited by load characteristics that inhibit platoon formation. Therefore, this research sets forth a suite of related truck platooning problems that correspond to alternative assumptions about unloaded-truck compensation and shipment temporal flexibility. The computational complexities of these problem variants are examined, [Formula: see text]-hardness is proved for the suite, and other useful mathematical properties are derived. Extensive empirical testing using the U.S. highway system explores the effect of network topology, commerce flow, and platooning labor savings on each problem’s optimal solution, tractability, and other related metrics. Multiple heuristics are also designed and tested for larger-scale instances. In so doing, this research not only explores the effects of varied factors on platooning labor costs but it also provides practical means to identify high-quality, labor-cost-reducing routes.

Rule-Based Engine for Automatic Allocation of Smallholder Dairy Producers in Preidentified Production Clusters.

Smallholder dairy producers account for around half of all African livestock ventures; nevertheless, they face challenges in producing more milk due to an insufficient framework and infrastructure to maximize their output. Smallholder dairy producers in this scenario use a variety of tactics to boost milk output. However, the attempts need multiple heuristics, time, and financial investment. Furthermore, because of a lack of extension officers, smallholder dairy producers become trapped in failure cycles, unsuccessful attempts, and a diminished motivation to continue farming. Therefore, the interventions were more straightforward as smallholder dairy producers with comparable characteristics grouped. This research aimed to create a rule-based engine that automatically assigns smallholder dairy producers to predefined clusters. About 78 stakeholders were interviewed, including 69 smallholder dairy producers and 9 extension officers from Meru-Arusha, Tanzania. The 10 production features and 6 predefined clusters were adopted from the previous study. Therefore, a rule-based engine used the selected 10 production features. As a result, the rule-based engine automatically assigns the smallholder dairy producers to their respective clusters. Therefore, smallholder dairy producers share their farming skills and experience to increase milk output through these clusters. Furthermore, extension officers in the system provide timely assistance to smallholder dairy producers with farming concerns.

Ant colony optimization equipped with an ensemble of heuristics through multi-criteria decision making: A case study in ensemble feature selection

Ant Colony Optimization (ACO) is a probabilistic and approximation metaheuristic algorithm to solve complex combinatorial optimization problems. ACO algorithm is inspired by the behavior of a colony of real ants and uses their pheromone trials to find optimal solutions. Since the beginning of the ACO algorithm, many researchers have tried to improve the performance and stability of the algorithm by using various methodologies. Resolving the exploitation/exploration dilemma by an efficient procedure is critical in improving the ACO. One of the critical parameters in ACO is selecting the heuristic that can affect the movements of ants. So far, the use of several heuristics in ACO has not been studied. We believe that using multiple heuristics instead of a single heuristic can improve the ACO algorithm. For this matter, we have proposed an ACO algorithm based on the ensemble of heuristics using a Multi-Criteria Decision-Making (MCDM) procedure. It means that the movement of the ants is defined based on the judgment of multiple experts (criteria). The idea is based on the hypothesis that different heuristics give us more information about the subsequent nodes, and the variety of these methods examines the different aspects to achieve better and optimal solutions in ACO. In this paper, we have applied our proposed method to the ensemble feature selection task to evaluate the performance of the proposed method. Blending several feature selection methods is regular to tackle the feature selection problem, and also efficiently combining feature selection methods is still challenging. Some well-known ensemble feature selection and primary feature selection methods have been compared with Ant-MCDM on twelve datasets to evaluate the performance of the proposed method in the feature selection task.

Quantifying behavioural difference in latent class models to assess empirical identifiability: Analytical development and application to multiple heuristics

Latent class (LC) models have been used for decades. In some cases, models of this kind have exhibited difficulties in identifying distinct classes. Identifiability is key to determining the presence or absence of the different population cohorts represented by the latent classes. Theoretical identifiability addresses this issue in general, but no empirical identifiability analysis of this kind of model has been performed previously. Here, we analyse the theoretical properties of LC models to establish necessary conditions on the classes to be identifiable jointly. We then, establish a measure of behavioural difference and relate it to empirical identifiability; this measure highlights factors that are crucial for identifiability. We show how these factors affect identifiability through simulation experiments in which classes are known, and test elements such as the proportion of individuals belonging to each latent class, different correlation structures and sample sizes. In our experiments, each latent class corresponds to a different choice heuristic. We present a graphical diagnostic that supports the measure of behavioural difference that promotes identifiability and provide examples of model non-identifiability, partial identifiability, and strong identifiability. We conclude by discussing how non-identifiability can be detected and understood in ways that will inform survey design and analysis.

Highly Accurate Real-Time Decomposition of Single Channel Intramuscular EMG.

Real-time intramuscular electromyography (iEMG) decomposition, as an identification procedure of individual motor neuron (MN) discharge timings from a streaming iEMG recording, has the potential to be used in human-machine interfacing. However, for these applications, the decomposition accuracy and speed of current approaches need to be improved. In our previous work, a real-time decomposition algorithm based on a Hidden Markov Model of EMG, using GPU-implemented Bayesian filter to estimate the spike trains of motor units (MU) and their action potentials (MUAPs), was proposed. In this paper, a substantially extended version of this algorithm that boosts the accuracy while maintaining real-time implementation, is introduced. Specifically, multiple heuristics that aim at resolving the problems leading to performance degradation, are applied to the original model. In addition, the recursive maximum likelihood (RML) estimator previously used to estimate the statistical parameters of the spike trains, is replaced by a linear regression (LR) estimator, which is computationally more efficient, in order to ensure real-time decomposition with the new heuristics. The algorithm was validated using twenty-one experimental iEMG signals acquired from the tibialis anterior muscle of five subjects by fine wire electrodes. All signals were decomposed in real time. The decomposition accuracy depended on the level of muscle activation and was when less than 10 MUs were identified, substantially exceeding previous real-time results. Single channel iEMG signals can be very accurately decomposed in real time with the proposed algorithm. The proposed highly accurate algorithm for single-channel iEMG decomposition has the potential of providing neural information on motor tasks for human interfacing.

Path Planning for Automatic Guided Vehicles (AGVs) Fusing MH-RRT with Improved TEB

In this paper, an AGV path planning method fusing multiple heuristics rapidly exploring random tree (MH-RRT) with an improved two-step Timed Elastic Band (TEB) is proposed. The modified RRT integrating multiple heuristics can search a safer, optimal and faster converge global path within a short time, and the improved TEB can optimize both path smoothness and path length. The method is composed of a global path planning procedure and a local path planning procedure, and the Receding Horizon Planning (RHP) strategy is adopted to fuse these two modules. Firstly, the MH-RRT is utilized to generate a state tree structure as prior knowledge, as well as the global path. Then, a receding horizon window is established to select the local goal point. On this basis, an improved two-step TEB is designed to optimize the local path if the current global path is feasible. Various simulations both on static and dynamic environments are conducted to clarify the performance of the proposed MH-RRT and the improved two-step TEB. Furthermore, real applicative experiments verified the effectiveness of the proposed approach.

Minimum k‐cores and the k‐core polytope

AbstractThe minimum k‐core problem asks for the smallest induced subgraph of minimum degree k. It has been shown that this problem is NP‐hard, and thus sophisticated techniques are required to obtain good solutions and approximations. In this article, the minimum k‐core problem is modeled as a binary integer program and relaxed as a linear program. Since the relaxation may yield a non‐integral solution, a branch‐and‐cut framework is used to find an integral optimal solution. It is shown that the edge and cycle transversals of the graph give valid inequalities for the convex hull of the k‐core polytope—which can be further generalized to a family of ‐core transversals. Further, a heuristic for the transversal of the minimal ‐cores is given with its associated valid inequality. Additionally, improved valid inequalities are generated using bounds involving the girth of the graph. Multiple heuristics are explored for finding initial bounds for the branching process utilizing the degree distribution of the graph. Finally, numerical results are given comparing the branch‐and‐bound, branch‐and‐cut, and heuristic techniques.

Improved Multi-Heuristic A* for Searching with Uncalibrated Heuristics

Recently, several researchers have brought forth the benefits of searching with multiple (and possibly inadmissible) heuristics, arguing how different heuristics could be independently useful in different parts of the state space. However, algorithms that use inadmissible heuristics in the traditional best-first sense, such as the recently developed Multi-Heuristic A* (MHA*), are subject to a crippling calibration problem: they prioritize nodes for expansion by additively combining the cost-to-come and the inadmissible heuristics even if those heuristics have no connection with the cost-to-go (e.g., the heuristics are uncalibrated) . For instance, if the inadmissible heuristic were an order of magnitude greater than the perfect heuristic, an algorithm like MHA* would simply reduce to a weighted A* search with one consistent heuristic. In this work, we introduce a general multi-heuristic search framework that solves the calibration problem and as a result a) facilitates the effective use of multiple uncalibrated inadmissible heuristics, and b) provides significantly better performance than MHA* whenever tighter sub-optimality bounds on solution quality are desired. Experimental evaluations on a complex full-body robotics motion planning problem and large sliding tile puzzles demonstrate the benefits of our framework.

Learning to Search More Efficiently from Experience: A Multi-heuristic Approach

Learning from experience can significantly improve the performance of search based planners, especially for challenging problems like high-dimensional planning. Experience Graph (E-Graph) is a recently developed framework that encodes experiences, obtained from solving instances in the past, into a single bounded-admissible heuristic, and uses it to guide the search. While the E-Graph approach was shown to be very useful for repetitive problems, it suffers from two issues. First, computing the E-Graph heuristic is time consuming as it maintains the bounded admissibility constraints. Second, a single heuristic can get stuck in a local minimum, and thereby, degrade the performance. In this work, we present an alternative approach to improving the runtime of search from experience, based on a recently developed search algorithm Multi-heuristic A* (MHA*). This framework provides an improvement over the E-Graph planner for two reasons: a) MHA* uses multiple heuristics simultaneously to explore the search space, which reduces the probability of getting stuck in a local minimum, and b) the heuristics in MHA* can be arbitrarily inadmissible, which makes it very easy to compute them. The paper describes the framework, explains how to compute these (inadmissible) heuristics through offline and online processing and presents experimental analysis on two domains, motion planning for a 6D planar arm and large sliding tile puzzles.

Towards Rational Deployment of Multiple Heuristics in A* (Extended Abstract)

In this paper we discuss and experiment with Lazy A*, a variant of A* where heuristics are evaluated lazily and with Rational Lazy A*, which decides whether to compute the more expensive heuristics at all, based on a myopic value of information estimate. Full version appears in IJCAI-2013.

Multidisciplinary design and optimization of expendable launch vehicle for microsatellite missions

PurposeThe capability to predict and evaluate various configurations’ performance during the conceptual design phase using multidisciplinary design analysis and optimization can significantly increase the preliminary design process’s efficiency and reduce design and development costs. This research paper aims to perform multidisciplinary design and optimization for an expendable microsatellite launch vehicle (MSLV) comprising three solid-propellant stages, capable of delivering micro-payloads in the low earth orbit. The methodology’s primary purpose is to increase the conceptual and preliminary design process’s efficiency by reducing both the design and development costs.Design/methodology/approachMultidiscipline feasible architecture is applied for the multidisciplinary design and optimization of an expendable MSLV at the conceptual level to accommodate interdisciplinary interactions during the optimization process. The multidisciplinary design and optimization framework developed and implemented in this research effort encompasses coupled analysis disciplines of vehicle geometry, mass calculations, aerodynamics, propulsion and trajectory. Nineteen design variables were selected to optimize expendable MSLV to launch a 100 kg satellite at an altitude of 600 km in the low earth orbit. Modern heuristic optimization methods such as genetic algorithm (GA), particle swarm optimization (PSO) and SA are applied and compared to obtain the optimal configurations. The initial population is created by passing the upper and lower bounds of design variables to the optimizer. The optimizer then searches for the best possible combination of design variables to obtain the objective function while satisfying the constraints.FindingsAll of the applied heuristic methods were able to optimize the design problem. Optimized design variables from these methods lie within the lower and upper bounds. This research successfully achieves the desired altitude and final injection velocity while satisfying all the constraints. In this research effort, multiple runs of heuristic algorithms reduce the fundamental stochastic error.Research limitations/implicationsThe use of multiple heuristics optimization methods such as GA, PSO and SA in the conceptual design phase owing to the exclusivity of their search approach provides a unique opportunity for exploration of the feasible design space and helps in obtaining alternative configurations capable of meeting the mission objectives, which is not possible when using any of the single optimization algorithm.Practical implicationsThe optimized configurations can be further used as baseline configurations in the microsatellite launch missions’ conceptual and preliminary design phases.Originality/valueSatellite launch vehicle design and optimization is a complex multidisciplinary problem, and it is dealt with effectively in the multidisciplinary design and optimization domain. It integrates several interlinked disciplines and gives the optimum result that satisfies these disciplines’ requirements. This research effort provides the multidisciplinary design and optimization-based simulation framework to predict and evaluate various expendable satellite launch vehicle configurations’ performance. This framework significantly increases the conceptual and preliminary design process’s efficiency by reducing design and development costs.

Flowshop scheduling with machine deterioration based on job sequences

This paper addresses the two-machine permutation flowshop problem with deterioration. The objectives are minimizing the makespan and the average tardiness. Jobs have a baseline process time in each machine and have a due date. The actual time to process a job depends on the machine performance level at the start of each job, which is a function of the previously processed jobs and their wear/deterioration effect on the machine. The article proposes multiple heuristics and a comprehensive set of experiments. The results indicate that as a group, the heuristics generate solutions that are very close to the optimal for both criteria. Furthermore, no heuristic approach is dominant for all experimental conditions, thus heuristic selection to solve practical problems should be based on the specific problem characteristics.

Learning Heuristic Selection with Dynamic Algorithm Configuration

A key challenge in satisficing planning is to use multiple heuristics within one heuristic search. An aggregation of multiple heuristic estimates, for example by taking the maximum, has the disadvantage that bad estimates of a single heuristic can negatively affect the whole search. Since the performance of a heuristic varies from instance to instance, approaches such as algorithm selection can be successfully applied. In addition, alternating between multiple heuristics during the search makes it possible to use all heuristics equally and improve performance. However, all these approaches ignore the internal search dynamics of a planning system, which can help to select the most useful heuristics for the current expansion step. We show that dynamic algorithm configuration can be used for dynamic heuristic selection which takes into account the internal search dynamics of a planning system. Furthermore, we prove that this approach generalizes over existing approaches and that it can exponentially improve the performance of the heuristic search. To learn dynamic heuristic selection, we propose an approach based on reinforcement learning and show empirically that domain-wise learned policies, which take the internal search dynamics of a planning system into account, can exceed existing approaches.

Subset-Saturated Transition Cost Partitioning

Cost partitioning admissibly combines the information from multiple heuristics for optimal state-space search. One of the strongest cost partitioning algorithms is saturated cost partitioning. It considers the heuristics in sequence and assigns to each heuristic the minimal fraction of the remaining costs that are needed for preserving all heuristic estimates. Saturated cost partitioning has recently been generalized in two directions: first, by allowing to use different costs for the transitions induced by the same operator, and second, by preserving the heuristic estimates for only a subset of states. In this work, we unify these two generalizations and show that the resulting subset-saturated transition cost partitioning algorithm usually yields stronger heuristics than the two generalizations by themselves.