Efficient Clustering Method for Graph Images Using Two-Stage Clustering Technique

Electronics, Journal Year: 2025, Volume and Issue: 14(6), P. 1232 - 1232

Published: March 20, 2025

Graphimages, which represent data structures through nodes and edges, present significant challenges for clustering due to their intricate topological properties. Traditional algorithms, such as K-means Density-Based Spatial Clustering of Applications with Noise (DBSCAN), often struggle effectively capture both spatial structural relationships within graph images. To overcome these limitations, we propose a novel two-stage approach that integrates conventional techniques graph-based methodologies enhance accuracy efficiency. In the first stage, distance- or density-based algorithm (e.g., DBSCAN) is applied generate initial cluster formations. second clusters are refined using spectral community detection better preserve exploit features. We evaluate our dataset 8118 images derived from depth measurements taken at various angles. The experimental results demonstrate method surpasses single-method approaches in terms silhouette score, Calinski-Harabasz index (CHI), modularity. score measures how similar an object its own compared other clusters, while CHI, also known Variance Ratio Criterion, evaluates quality based on ratio between-cluster dispersion within-cluster dispersion. Modularity, metric commonly used clustering, assesses strength division network into communities. Furthermore, qualitative analysis visualization confirms proposed more differentiates similarities These findings underscore potential hybrid applications, including three-dimensional (3D) measurement analysis, medical imaging, social analysis.

Language: Английский

---

Joint Approach for Vehicle Routing Problems Based on Genetic Algorithm and Graph Convolutional Network

Mathematics, Journal Year: 2024, Volume and Issue: 12(19), P. 3144 - 3144

Published: Oct. 8, 2024

The logistics demands of industries represented by e-commerce have experienced explosive growth in recent years. Vehicle path-planning plays a crucial role optimization systems for and distribution. A scheme suitable an actual scenario is the key to reducing costs improving service efficiency industries. In complex application scenarios, however, it difficult conventional heuristic algorithms ensure quality solutions vehicle routing problems. This study proposes joint approach based on genetic algorithm graph convolutional network solving capacitated problem with multiple distribution centers. First, we use method modularize environment encode each module constraint conditions. Next, adopted feature embedding representation problem, decoders are used increase diversity solution space. Meanwhile, REINFORCE baseline employed train model, ensuring quick returns high-quality solutions. Moreover, fitness function calculated module, seek optimal global scale. Finally, effectiveness proposed framework validated through experiments at different scales comparisons other algorithms. experimental results show that, compared single decoder GCN-based method, this paper improves success rate 100% across 15 generated instances. average path length obtained only 11% produced multi-decoder method.

Language: Английский

---

A review on sustainability, Industry 4.0 and collaboration implications in vehicle allocation operations

The International Journal of Logistics Management, Journal Year: 2023, Volume and Issue: 35(3), P. 943 - 978

Published: Dec. 9, 2023

Purpose Vehicle allocation problems (VAPs), which are frequently confronted in many transportation activities, primarily including but not limited to full truckload freight operations, induce a significant economic impact. Despite the increasing academic attention field, literature still fails match needs of and opportunities growing industrial practices. In particular, can grow upon ideas on sustainability, Industry 4.0 collaboration, shape future practices only logistics also other industries. This review has potential enhance accelerate development relevant that matches challenges problems. Furthermore, this help explore existing methods, algorithms techniques employed address problem, reveal directions generate inspiration for improvements. Design/methodology/approach study provides VAPs, focusing quantitative models incorporate any following emerging trends: collaboration. Findings literature, sustainability interactions have been environmental externalities (mostly reducing operational-level emissions) considerations; however, emissions generated throughout supply chain, such as waste product deterioration, or level stakeholder engagement, etc., be monitored order achieve overall climate-neutral services society. Moreover, even though there types collaboration (such co-opetition vertical collaboration) sharing information comanaging distribution operations) could improve vehicle these topics yet received sufficient from researchers. Originality/value The scientific contribution is twofold: (1) analyses decision each reviewed article terms variable, constraint assumption sets, objectives, modeling solving approaches, way aspects incorporated into model. (2) authors provide discussion gaps related particularly practical serving climate-neutrality targets, carried out under four main streams: possibilities, chain risks, smart solutions various As result, several and/or research may positive impacts, how further engaged, risks models, cope with uncertainty effectiveness efficiency operations.

Language: Английский

---

Scheduling in services: a review and bibliometric analysis

Operations Management Research, Journal Year: 2024, Volume and Issue: 17(2), P. 754 - 783

Published: March 11, 2024

Language: Английский

---

Visualization and Analysis of Hotspots and Trends in Seafood Cold Chain Logistics Based on CiteSpace, VOSviewer, and RStudio Bibliometrix

Sustainability, Journal Year: 2024, Volume and Issue: 16(15), P. 6502 - 6502

Published: July 30, 2024

The development of cold chain logistics for seafood plays a pivotal role in guaranteeing food safety, promoting economic progress, reducing losses, and fostering sustainable development, thereby enhancing the overall efficiency supply chain. This study conducted comprehensive investigation into primary research focuses field using literature on visualization analysis software (CiteSpace (6.2.R6), VOSviewer (1.6.20), RStudio Bibliometrix (4.4.0)). A total 1787 articles were collected further analyzed from China National Knowledge Infrastructure (CNKI), Web Science (WOS), Google databases over 12 years, establishing knowledge framework logistics. Through utilization keyword clustering emerging techniques, constructed map that intuitively describes trends key hotspot this field. results indicate growing trend field, with disciplines such as mathematics, systems, physics being notably prominent. Key terms “cold logistics”, “highlighted management”, “frozen storage techniques”, practices”, “post-harvest loss prevention strategies”, “optimization chain” frequently appear literature, highlighting importance interdisciplinary academic these areas. By exploring current we strengthen gaps propose future directions. Therefore, well-conducted bibliometric studies can play crucial advancing by providing insights, facilitating scholarly discussions, identifying gaps, generating new ideas, showcasing their intended contributions to

Language: Английский

---

An Improved Ant Colony Algorithm with Deep Reinforcement Learning for the Robust Multiobjective AGV Routing Problem in Assembly Workshops

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(16), P. 7135 - 7135

Published: Aug. 14, 2024

Vehicle routing problems (VRPs) are challenging problems. Many variants of the VRP have been proposed. However, few studies on combined robustness and just-in-time (JIT) requirements with uncertainty. To solve problem, this paper proposes just-in-time-based robust multiobjective vehicle problem time windows (JIT-RMOVRPTW) for assembly workshop. Based conflict between uncertain JIT requirements, a strategy was measure solution, metric designed as objective. Afterwards, two-stage nondominated sorting ant colony algorithm deep reinforcement learning (NSACOWDRL) In stage I, ACO combines NSGA-III to obtain Pareto frontier. model, pheromone update transfer probability formula were designed. DDQN introduced local search which trains networks through solutions participate in probabilistic selection sorting. II, frontier quantified feasibility by Monte Carlo simulation, tested diversity-robust based uniformly distributed weights solution space select that take diversity into account. The effectiveness NSACOWDRL demonstrated comparative experiments other algorithms instances. impact is analyzed effect further discussed.

Language: Английский

---

Vehicle Route Planning of Diverse Cargo Types in Urban Logistics Based on Enhanced Ant Colony Optimization

World Electric Vehicle Journal, Journal Year: 2024, Volume and Issue: 15(9), P. 405 - 405

Published: Sept. 4, 2024

In the realm of urban logistics, optimizing vehicle routes for varied cargo types—including refrigerated, fragile, and standard cargo—poses significant challenges amid complex infrastructures heterogeneous capacities. This research paper introduces a novel model multi-type capacitated routing problem (MT-CVRP) that harnesses an advanced ant colony optimization algorithm, dubbed Lévy-EGACO. algorithm integrates Lévy flights elitist guiding principles, enhancing search efficacy pheromone update processes. The primary objective this study is to minimize overall transportation costs while efficiency intricate route planning vehicles with diverse load Through rigorous simulation experiments, we corroborated validity proposed effectiveness Lévy-EGACO in routes. demonstrated consistent reduction costs, ranging from 1.8% 2.5% compared other algorithms, across different test scenarios following base data modifications. These findings reveal substantially improves optimization, presenting robust solution MT-CVRP within logistics frameworks.

Language: Английский

---

An exact method for vehicle routing problem with backhaul discounts in urban express delivery network

Cleaner Logistics and Supply Chain, Journal Year: 2024, Volume and Issue: 11, P. 100157 - 100157

Published: June 1, 2024

The surge in e-commerce has led to an increased demand for urban express services, requiring the strategic development of delivery networks that are both efficient and cost-effective. This study addresses a practical vehicle routing problem (VRP) network minimize transportation costs. Specifically, it considers implementation backhaul discounts, factor disregarded existing literature. VRP is further complicated by various realistic constraints, including pickup delivery, time windows, multiple trips, heterogeneous fleets, docking capacity limitations, which make most general solvers inapplicable. proposes trip-based formulation overcome this challenge develop tailored branch-and-price algorithm. Feasible trips classified into four types simplify computation thereby enhancing solution efficiency. Validation with real-world data from SF Express substantiates efficacy our method yields insights sustainable city logistics management. Moreover, simplified column generation algorithm exhibits competitive performance, achieving optimal solutions expeditiously tested instances.

Language: Английский

---

Efficient Expansion Algorithm of Urban Logistics Network for Medical Products Considering Environmental Impact

Sustainability, Journal Year: 2024, Volume and Issue: 16(10), P. 4195 - 4195

Published: May 16, 2024

As society continues to age, people are becoming more concerned about their health care. This has led an increase in the demand for medical products urban areas, emphasizing need regular and prompt deliveries. However, existing logistics centers located suburbs of Seoul, a metropolitan city, which makes it challenging ensure timely delivery. To address this issue, study aims establish new particularly while minimizing CO2 emissions from delivery vehicles alignment with sustainability efforts. The scientific gap addressed novelty paper is that input parameters prepared based on actual data company Korea reflect reality, mathematical model-based optimization technique applied determine optimal location center. genetic algorithm developed solve proposed model by deriving or near-optimal solutions. Furthermore, numerical experiment examined impact establishing center one candidate areas local governments Seoul considering environmental impact. result, network can reduce approximately 66.74% compared network.

Language: Английский

---

ANT ALGORITHM APPLICATION FOR ROUTE PLANNING FOR UNMANNED AERIAL VEHICLES (UAVs)

Наука і техніка сьогодні, Journal Year: 2024, Volume and Issue: 7(35)

Published: Aug. 2, 2024

Creating routes for drones is a convoluted process hence this objective requires to design optimal itineraries safe interaction between groups of remotely piloted aerial systems (RPAS).Also, drone exploitation opens up completely new perspectives in logistic control because, comparison with ground vehicle, RPAS uses airspace as its primary road flying directly the designated point.However, it must be maintained that such small aircraft vulnerable weather conditions and does not have an excessive amount energy which operator always has keep mind when he intends take off.Thereby, problem route consists numerous restrictions demands algorithms efficient navigation.The article devoted actual using unmanned vehicles human activity.In particular, considers task routing vehicles, complex optimisation involving determination group UAVs perform various tasks.The subject environment characterised, may include following components: objects recharging stations.It describes several existing analogues platforms can used planning particular DJI Terra Altitude Angel.The authors consider most common types vehicle problems (Capacitated Vehicle Routing Problem, VRP Time Windows, Multiple Depot VRP).The paper formulates mathematical statement UAV problem.№ 7(35) 2024 502Considerable attention paid ant algorithm modifications.The define importance prospects further research on use heuristic controlling semi-automated systems.

Language: Английский

---