Constructive heuristics for the Mixed Capacity Arc Routing Problem under Time Restrictions with Intermediate Facilities

In the emerging Internet of Things (IoT) paradigm, the need for real-time decision support is increasing fast to support key applications, such as smart health, disaster recovery, or battlefield monitoring. In these applications, it is essential to efficiently retrieve fresh sensor data and process them in a timely manner to aid in decision making. However, related work is relatively scarce. In this paper, we formulate the problem of real-time decision support and explore a suite of scheduling methodologies to efficiently retrieve and process sensor data for real-time decision support subject to timing and data freshness constraints at the network edge and discuss open research issues.

This work explores the concept of improvisation as a framework for understanding real-time dynamic decision making (RTDDM) and systems support for it. The contexts for RTDDM and those in which agents improvise are remarkably similar according to several parameters. The foundations of improvisation are built on management theory, practice, and education. The work provides a rich definition of improvisation and a typology of different improvisational contexts based on two key dimensions. This framework illustrates how seemingly diverse contexts such as emergency management and jazz performance are related. The work then explores the antecedents of improvisation, degrees of improvisation, the elements for individual and team improvisation, and effective improvisation. The work then explores the design of Real-Time dynamic decision support systems (DSS). These are broken into pre- and postperformance support, as well as Real-Time support. The conclusion is that our knowledge of improvisational contexts can shed new light on RTDDM systems design and development.

Efficient route optimization is critical for municipal solid waste collection, as poorly planned routes increase operational costs, environmental impact, and resource consumption. This study compares three path optimization algorithms-Ant Colony Optimization, Dijkstra's algorithm, and the Nearest Neighbour heuristic-in the context of municipal waste collection networks. The algorithms are evaluated on simulated graphs of varied sizes to assess path cost, computational time, and solution stability. Ant Colony Optimization demonstrates higher adaptability, cost-effectiveness, while Dijkstra provides deterministic optimal pathways with lower variance. The Nearest Neighbour approach, although computationally faster, constantly produces inferior routes. Performance disparities were statistically validated using the Wilcoxon signed-rank test. The findings offer practical guidelines for selecting efficient routing algorithms in urban waste collection systems and establish a foundation for future improvements incorporating real-time data and hybrid models.

GHG emission factors developed for the collection, transport and landfilling of municipal waste in South African municipalities

The sectoring arc routing problem (SARP) is introduced to model activities associated with the streets of large urban areas, like municipal waste collection. The aim is to partition the street network into a given number of sectors and to build a set of vehicle trips in each sector, to minimize the total duration of the trips. Two two-phase heuristics and one best insertion method are proposed. In the two-phase methods, phase 1 constructs the sectors using two possible heuristics, while phase 2 solves a mixed capacitated arc routing problem (MCARP) to compute the trips in each sector. The best insertion method determines sectors and trips simultaneously. In addition to solution cost, some evaluation criteria such as imbalance, diameter and dispersion measures are used to compare algorithms. Numerical results on large instances with up to 401 nodes and 1056 links (arcs or edges) are reported and analysed.

Real-time clinical decision support (CDS) integrated with anesthesia information management systems (AIMS) can generate point of care reminders to improve quality of care. To develop, implement and evaluate a real-time clinical decision support system for anesthetic management of pediatric traumatic brain injury (TBI) patients undergoing urgent neurosurgery. We iteratively developed a CDS system for pediatric TBI patients undergoing urgent neurosurgery. The system automatically detects eligible cases and evidence-based key performance indicators (KPIs). Unwanted clinical events trigger and display real-time messages on the AIMS computer screen. Main outcomes were feasibility of detecting eligible cases and KPIs, and user acceptance. The CDS system was triggered in 22 out of 28 (79%) patients. The sensitivity of detecting continuously sampled KPIs reached 93.8%. For intermittently sampled KPIs, sensitivity and specificity reached 90.9% and 100%, respectively. 88% of providers reported that CDS helped with TBI anesthesia care. CDS implementation is feasible and acceptable with a high rate of case capture and appropriate generation of alert and guidance messages for TBI anesthesia care.

The rapid development of online shopping has led to an increase in express parcels, thus the proliferation of solid waste from parcel packaging. There are many other phenomena that have arisen with rapid economic development that also led to the surge in municipal solid waste. It is urgent to pay attention to municipal solid waste collection. Unexpected disruptive events lead to cognitive uncertainty and cause municipal waste collection to fail to function properly. This article studies how to properly carry out vehicle planning to achieve the shortest driving distance under cognitive uncertainty. The community's time window and the amount of garbage in each community are taken as uncertain variables. Based on the uncertainty theory, an uncertain vehicle routing problem (VRP) model and an uncertain VRP with the loading constraint model are constructed, and the solution algorithms of these two models are given. Several applications are carried out to evaluate the effectiveness of the proposed models and algorithms. This article proposes for the first time the models and algorithms to study the VRP under cognitive uncertainty. The findings draw important practical implications for adjusting municipal solid waste collection planning and logistics during disruptive events.

Technological advances in threat technology, the increasing tempo and diversity of open-ocean and littoral scenarios, and the volume and imperfect nature of data to be processed under time-critical conditions pose significant challenges for future shipboard Command and Control Systems and operators who must use these systems to defend their ship and fulfil their mission. To address these challenges, we are investigating design concepts for a real-time decision support system that continuously fuses data from the ship's sensors and other sources, helps operators maintain a tactical situation picture, and supports their response to actual or anticipated threats. This paper is concerned with support for weapon engagement management in above-water warfare that forms part of the system. The manager plans and directs in real time actions involving use of hardkill and softkill weapons to counter air and surface threats. We propose an agent architecture that dynamically interleaves planning and execution, as it adapts to unanticipated changes in the tactical situation. The architecture integrates both reactive and deliberative planning. Deliberation is based on projecting the currently perceived situation and may incorporate plan contingencies to handle action outcome and situation uncertainty. Reactive planning is driven by precompiled stimulus-response knowledge (e.g., standard operating procedures). Local control of architectural components to allow for varying computational resources and time criticality for producing deliberative plans is also briefly examined. Keywords: weapon engagement management, command and control system, real-time decision support, agent architecture, planning.

This research presents a solution process for Periodic Arc Routing Problem with Refill Points (PARPRP). The Periodic Arc Routing Problem is a natural extension of Arc Routing Problem (ARP) and is classified as an NP-hard problem. The street tree watering problem is used as the case study. The study tackles the ARP requiring a specified time horizon to fulfill it. A refill point in the problem is a node in the graph at which the vehicle capacity can be recovered as long as the vehicle makes a stop on it. The strategy of solving the problem is first to transform the problem into a Vehicle Routing Problem. Secondly, an Ant Colony Optimization algorithm is developed to solve the PARPRP because of its remarkable capability in the establishment of network routes. A local search method is incorporated to reach the local minimum for further improvement. Two sets of benchmarks are used to verify the solving process and the computational results are provided. The outputs indicate that proposed methods can yield promising solutions within reasonable execution time. A real case regarding the street tree watering of Kaohsiung city, Taiwan, is investigated. The results suggest that while α = 1 and β = 3 (i.e. parameters to control the influence of the pheromone value allocated on and the desirability of the arc, respectively), the algorithm can yield the best solution. The number of refill points is analyzed as well and output result suggests that a system with two refill points can obtain the least route length.

This paper presents a decision support for addressing urban waste collection services that are characterized by the presence of a garage from which all the vehicles start their routes, an intermediate facility where the vehicles unload the waste and a large number of pick up positions. Moreover, in a shift the vehicle can perform several routes subject to time and capacity constraints. Our aim is supporting the decision maker in planning the optimal vehicle routes with the objective of minimizing the length of the travels while collecting all the daily waste. The Waste Collection Vehicle Routing Problem (WCVRP) is modelled as a Mixed Integer Linear Programming problem with shift and time constraints. In order to address the large dimension of the considered city street networks, we propose a heuristic algorithm based on a clustering strategy and a farthest insertion heuristic for the solution of the Traveling Salesman Problem. A set of tests and a case study show the effectiveness of the proposed heuristic algorithm to plan the waste collection in a work day of an Italian city.

Explainable Artificial Intelligence (XAI) has emerged as a critical component in modern real-time decision support systems, where transparency, trust, and accountability are essential. Traditional AI models, particularly deep learning techniques, often function as “black boxes,” making it difficult for users to understand how decisions are derived. This research paper explores the design, implementation, and evaluation of explainable AI models tailored for real-time decision support environments. The study focuses on integrating interpretable machine learning techniques, such as feature importance analysis, rule-based explanations, and model-agnostic explanation methods, into time-sensitive applications. It also examines the trade-off between model accuracy and interpretability in dynamic decision-making scenarios such as healthcare monitoring, financial risk assessment, and smart industrial systems. Furthermore, the paper proposes a hybrid explainable framework that combines high-performance predictive models with post-hoc explanation tools to ensure both efficiency and transparency. Experimental analysis demonstrates that the proposed approach enhances user trust, improves decision quality, and enables faster human–AI collaboration without significantly compromising system performance. The findings highlight the importance of explainability as a foundational requirement for deploying AI-driven real-time decision support systems in critical domains.

Anthrax is a potential biological weapon and can be used in an air-borne or mail attack, such as in the attack in the United States in 2001. Planning for such an event requires the best available science. Since large-scale experiments are not feasible, mathematical modelling is a crucial tool to inform planning. The aim of this study is to systematically review and evaluate the approaches to mathematical modelling of inhalational anthrax attack to support public health decision making and response. A systematic review of inhalational anthrax attack models was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria. The models were reviewed based on a set of defined criteria, including the inclusion of atmospheric dispersion component and capacity for real-time decision support. Of 13 mathematical modelling studies of human inhalational anthrax attacks, there were six studies that took atmospheric dispersion of anthrax spores into account. Further, only two modelling studies had potential utility for real-time decision support, and only one model was validated using real data. The limited modelling studies available use widely varying methods, assumptions, and data. Estimation of attack size using different models may be quite different, and is likely to be under-estimated by models which do not consider weather conditions. Validation with available data is crucial and may improve models. Further, there is a need for both complex models that can provide accurate atmospheric dispersion modelling, as well as for simpler modelling tools that provide real-time decision support for epidemic response.

A Waste Collection Arc Vehicle Routing Problem with turn constraints (WCAVRPTC) model is proposed to minimize the length of municipal solid waste (MSW) collection routes, which presents several characteristics derived from traffic regulations: some streets can be traversed crossed in only one direction, and some turns should be forbidden when the common angle is too narrow. The problem is transformed into a node routing ones and a clustering-based multiple ant colony system algorithm is described to address it. This algorithm possesses a set of new strategies to construct solutions, including clustering-based strategy, improved global pheromone update rule. The methodology is also applied to the urban waste collection problem in Shuangnan. The results of optimization indicate that the length of saving routes is 89984.96m and 92330.04m respectively with or without turns constrains constraints, which decreased by 31.1% and 31.9% respectively.

Collection of Municipal Solid Waste (MSW) is important in every waste management program. Communal container collection systems appear most prevalent in many developing countries. However, this collection system is associated with problems such as overflow of waste containers, ground dumping at collection sites, and open/indiscriminate dumping at unauthorized places. The spatial distribution of these activities present potential contamination challenges to water resources. Spatial information on Municipal Solid Waste Collection/Dump Sites (MSWCS) is essential for Municipal Solid Waste Management (MSWM) decision-making, including siting and collection route planning, and dumps cleanup. Integration of Geographic Information Systems (GIS) and Global Positioning Systems (GPS) present a platform to capture, map, and analyze spatial MSWM issues. This study applied GIS and GPS technologies to map MSWCS in Wa town, Ghana. Wa Municipality waste collection system efficiency was analyzed based on spatial availability of communal containers. A GIS model was developed and used to analyze proximity of MSWCS to boreholes drinking water sources. A pollution risk map was also developed to analyze the potential impact of indiscriminate dumps on hand-dug wells and surface water in Wa. The results reveal poor management in a significant number (about 67%) of MSWCS sites. Fifteen percent of mechanized boreholes failed the minimum proximity requirement (≥100 m) of the model. This study findings illustrate that MSWM in developing countries may be improved using GIS and GPS technologies. The approach used for Wa could be replicated in other towns in Ghana and other developing nations with similar MSWM problems to enhance policy and management decision-making.

Is bigger better? An empirical analysis of waste management in New South Wales