The researchers’ starting point is the ant colony system, a family of algorithms inspired by the way real ants find short paths to food. Individual ants deposit pheromone along the trails they walk, and subsequent ants preferentially follow stronger trails, reinforcing good routes while weaker ones evaporate. Translated into computation, artificial ants construct delivery routes step by step, guided by two signals: pheromone trails that encode collective experience about which customer-to-customer moves have historically produced good routes, and heuristic information that estimates the immediate desirability of each possible move. Over many iterations, the colony converges toward high-quality solutions without ever enumerating the full search space. Since Marco Dorigo introduced ant colony optimization in the early 1990s, the approach has been applied to everything from network routing to scheduling, and it remains a workhorse for routing problems where exact methods run out of steam.

What distinguishes the new algorithm is the way it deploys not one but four heuristic techniques inside the colony’s decision-making machinery. In a conventional ant colony system, each ant typically relies on a single heuristic rule when choosing the next customer to visit. That rigidity can be a liability, because different heuristics excel in different regions of the search: one might be better at deciding which customer to add to a route, another at deciding where to insert a customer into an existing route, and still others at repairing or restructuring routes that have become inefficient. Rather than betting on a single rule, the authors let the algorithm learn which heuristic serves it best at any given moment, drawing on the complementary strengths of all four.

The mechanism that orchestrates this cooperation is a ranking-based selection procedure built on the classic roulette wheel. Each of the four heuristics is assigned a weight reflecting how well it has performed recently, and when the algorithm needs to make a decision, it spins the metaphorical wheel, giving better-ranked heuristics proportionally larger slices. Crucially, the weights are not fixed. As the search progresses, heuristics that consistently contribute to improved solutions gain a larger share of future decisions, while those that fail to help are gradually demoted. This adaptive allocation means the algorithm effectively conducts a continuous, built-in experiment, reallocating its computational effort toward whichever strategy is currently paying off. The result, according to the authors, is the generation of promising solutions that no single heuristic could reliably produce alone.

A second design element addresses one of the most persistent failure modes of swarm intelligence: stagnation. Because pheromone reinforcement favors routes that have already been found good, a colony can settle into a rut, repeatedly exploring the same small set of routes while better alternatives elsewhere in the search space go unnoticed. The new algorithm counters this by incorporating a no-improvement counter into its solution selection process. When the counter indicates that the colony has gone a stretch of iterations without finding anything better, the algorithm adjusts its selection behavior to push the search toward less-explored territory, restoring diversity among the candidate solutions. This balance between intensification, which exploits known good regions, and diversification, which explores new ones, is the central tension of all metaheuristic design, and the counter gives the algorithm an explicit, self-monitoring handle on it.

The third and perhaps most inventive contribution is a new local improvement operator the authors call 2-opt-special, inspired by the venerable 2-opt move that has been a staple of routing heuristics since the 1960s. In its classic form, 2-opt takes a route, removes two edges, and reconnects the remaining segments in the alternative way, undoing any crossings and shortening the tour. The special variant introduced here adapts and extends this idea to bolster both the efficiency and the effectiveness of the ant colony framework, giving the colony a targeted tool for refining the routes its ants construct. Local search operators of this kind matter enormously in practice: constructive methods such as ant colonies are good at assembling plausible routes, but the difference between a plausible route and an excellent one often lies in small, surgical repairs that only a well-designed neighborhood move can deliver.

To find out whether all this machinery actually works, the team ran extensive computational experiments on well-known benchmark instances containing 100 and 400 customers, the standard proving grounds for VRPTW algorithms. They compared their method against other ant colony-based algorithms and against established solutions from the literature. The results, the authors report, affirm that the proposed algorithm is effective and competitive in addressing VRPTW challenges, holding its own against prior approaches across the benchmark suite. The comparisons were supported by the kind of statistical methodology that has become standard practice for evaluating evolutionary and swarm algorithms, following nonparametric testing procedures designed to compare stochastic optimizers fairly rather than relying on single lucky runs.

The significance of the work extends beyond the leaderboard. Vehicle routing with time windows sits at the heart of logistics, e-commerce, and emergency transportation, and even percentage-point improvements in route quality translate into real savings in fuel, fleet size, and emissions. The paper’s emphasis on combining multiple heuristics under an adaptive selection scheme reflects a broader trend in metaheuristics research: rather than designing ever more exotic single algorithms, researchers increasingly build frameworks that can assemble and tune proven components on the fly. The authors’ own research trajectory points in the same direction, including earlier work on hybrid adaptive large neighborhood search for capacitated routing problems and on metaheuristic techniques applied to engineering design problems.

For the field, the study also offers a reminder of how much headroom remains in a technique that is more than three decades old. Ant colony optimization was conceived as a simple model of distributed learning, yet its modern descendants now incorporate ranking mechanisms, adaptive strategy portfolios, stagnation detection, and specialized local search, layering decades of algorithmic insight onto the original biological metaphor. The version of the algorithm described here, published in volume 38 of Neural Computing and Applications as article number 705, was received in October 2024 and accepted in July 2026, and the authors declare no conflicts of interest and no specific funding for the work.

As delivery networks grow denser and customer expectations for precise delivery windows tighten, the computational problems underlying logistics will only intensify. Approaches like the one developed by Kuyu, Fajardo-Calderin, and Onieva suggest that the path forward lies not in any single clever rule but in flexible systems that know when to switch tactics, when to double down on what works, and when to break out of a rut. Virtual ants, it turns out, still have plenty to teach us about finding the best way home.

Subject of Research: A novel ant colony system using multiple heuristic strategies for solving the vehicle routing problem with time windows

Article Title: A novel ant colony system for vehicle routing problem with time windows: leveraging multiple heuristic strategies for improved solutions

Article References: Kuyu, Y. Ç., Fajardo-Calderin, J., & Onieva, E. (2026). A novel ant colony system for vehicle routing problem with time windows: leveraging multiple heuristic strategies for improved solutions. Neural Computing and Applications, 38(17), Article 705. https://doi.org/10.1007/s00521-026-12418-z

Image Credits: AI Generated

DOI: 10.1007/s00521-026-12418-z

Keywords: ant colony optimization, vehicle routing problem, time windows, heuristics, metaheuristics, swarm intelligence, combinatorial optimization, logistics, local search, 2-opt, benchmark instances, Neural Computing and Applications

The stakes are both environmental and economic. Decommissioned crystalline silicon modules contain lead, cadmium and tin, substances linked to brain disorders, cancers and damage to the lungs, liver, kidneys and prostate if they leach from improper dumps. Yet the same panels are also a treasure chest of recoverable material. IRENA estimates that recycling end-of-life PV panels could unlock roughly 450 million dollars in recoverable value by 2030 and 15 billion dollars by 2050. The problem, the study’s authors Shikha Tiwari, Eshwar Dayal and Lakshay argue, is that existing waste management systems simply fail to capture this stream, because of economic and logistical constraints. Solar installations range from sprawling utility-scale parks to scattered urban rooftops, and that geographic fragmentation drives up the cost of collecting heavy, fragile freight from thousands of dispersed points.

The core of the new work is a bi-level optimization model, a mathematical structure designed for situations where two decision-makers with conflicting goals interact. In this formulation, the collection centre operator acts as the leader, seeking to minimise total cost, which includes fixed facility costs, vehicle routing expenses and the incentives paid out. The end users of solar panels act as followers, rationally choosing whether to hand their retired panels to the formal recycling system and, if so, whether to request a home pickup or deliver the panels themselves to a drop-off point. The lower level of the model maximises the incentives users receive, while the upper level minimises the operator’s total cost. The authors frame this as a Stackelberg game, a classic game-theoretic construct in which the leader anticipates the follower’s best response before committing to its own decision.

What makes the framework distinctive is that it stitches together, in a single model, four elements that previous studies treated largely in isolation: forecasting of future waste volumes, optimisation of facility locations, vehicle routing, and financial incentives. The authors’ comparative analysis of the literature shows that most prior works incorporated at most two of these dimensions. Some optimised facility placement but ignored routing; others modelled incentives but lacked any mathematical backbone for reverse logistics. By contrast, the new model makes incentives endogenous, meaning the payment levels themselves are decision variables that respond to waste volumes and costs, rather than fixed assumptions plugged in from outside.

The forecasting component is built on the Weibull distribution, a statistical function widely used to model failure rates over time. Rather than assuming every panel dies at exactly year 25, the Weibull approach produces a probabilistic retirement curve that rises during operational degradation and accelerates near end of life, which the authors argue is far closer to real-world behaviour than a fixed-lifespan model in which all panels retire simultaneously. The study runs two scenarios: a regular-loss baseline with a shape factor of about 2.49, and an early-loss pessimistic scenario with a shape factor of 5.38, reflecting premature failures from harsh weather or handling damage. Future installed capacity is projected using segmented linear growth rates, declining from 7.8 percent in 2021-2025 to 3.5 percent by 2036-2040, a deliberate choice to avoid the overestimation that exponential extrapolation produces over a 20-year horizon.

That overestimation problem is vividly demonstrated in the results. When an exponential growth assumption with a 14 percent compound annual growth rate is applied to Jaipur district, the model projects more than 2,300 megawatts of installed capacity in a single district by 2040, and a cumulative capacity overestimate of more than 100 percent. Such inflated numbers would lead planners to open unnecessary collection centres, oversize their fleets and set incentive levels for waste that will never materialise. The segmented linear approach, by contrast, produces projections that the authors consider implementable, and it feeds district-level waste estimates for Jaipur from 2021 to 2040, a spatial resolution fine enough to support actual facility location and routing decisions, unlike the national or state-level estimates that dominate the existing literature.

For the case study, Jaipur, the district with the highest installed solar capacity in India’s top solar state, was divided into 95 equal square grids, with each grid centroid serving as a waste generation site. Thirty-seven authorised e-waste collection centres provided the pool of candidate facilities. The routing side of the model is a capacitated vehicle routing problem, solved with Miller-Tucker-Zemlin subtour elimination constraints to ensure that every collection route is a single continuous loop returning to its depot, rather than a set of disconnected fragments. Haversine distances between grid centroids and centres ground the routing calculations in real geography. The model also enforces a profitability rule: revenue from selling collected panels to recyclers must exceed total operating cost by at least 10 percent, a threshold reflecting administrative overhead, price volatility in recovered materials and the need to fund infrastructure investment.

The results reveal an elegant adaptive behaviour. Under regular loss, the model keeps just two collection centres open until 2035, then adds one per year, reaching six by 2039. Under early loss, where cumulative waste over 2021-2030 is roughly 31 times higher than in the regular scenario, the network scales to as many as 14 centres. Crucially, the model remains profitable in every year of both scenarios, adjusting centre counts and incentive levels without any structural redesign. The incentive dynamics are particularly telling: at low waste volumes, pickup incentives sit near their upper limit because door-to-door collection is cheap when there is little to collect, but as volumes surge, the model widens the gap between drop-off and pickup payments, effectively shifting the logistical burden onto end users and encouraging them to deliver panels themselves. Incentive values eventually stabilise at around 2,500 rupees per kilowatt for pickup and 3,500 for drop-off.

The study also derives four key performance indicators that translate the mathematics into measurable outcomes: cost per tonne collected, average haul distance per tonne, waste recovery rate, and carbon dioxide emissions per tonne. The recovery rate holds steady at about 84 percent across all years and scenarios, while emissions per tonne fall dramatically as waste density rises, dropping from over 35 kilograms of CO2 per tonne in the sparse early years to under 2 kilograms by 2039 in the regular-loss case. Sensitivity analysis shows that minimum collection rates above 70 percent yield a healthy profit-to-cost ratio, that collection centre capacities of 40,000 kilograms or more stabilise the network at two facilities, and that vehicle capacities of 4,000 kilograms or more minimise cost. Computationally, the framework scales gracefully: even at 132 grid cells, generating over four million variables, the solver converged in under 42 seconds with optimality gaps below 2.5 percent.

The implications reach well beyond Jaipur. The authors recommend that policymakers provide subsidies or extended producer responsibility funding during the low-volume window of 2026 to 2030, explicitly incentivise self-drop-off since full home pickup is fiscally unviable, and use the model’s quantified costs to set collection targets that are both ambitious and financially realistic. For operators, the message is that flat-rate incentives are inefficient; payments should be dynamically adjusted, centres should open in stages, and the pickup-drop-off gap should widen during waste surges. The framework is designed to be replicated across regions and time frames, and with modifications could extend to other e-waste streams. Future work, the authors note, could incorporate stochastic user behaviour, the informal recycling sector as a competitor, and real-time GIS integration. But the central insight stands: the coming wave of solar waste is not merely a liability to be buried, but a feedstock whose recovery can be engineered, optimised and made to pay for itself.

Subject of Research: Bi-level optimization modeling for sustainable end-of-life solar photovoltaic waste management and reverse logistics within a circular economy framework

Article Title: Development of a bi-level optimization model for sustainable Solar PV Waste management within a circular economy framework

Article References: Tiwari, S., Dayal, E., & Lakshay (2026). Development of a bi-level optimization model for sustainable Solar PV Waste management within a circular economy framework. Cleaner Engineering and Technology, 35, Article 101327. https://doi.org/10.1016/j.clet.2026.101327

Image Credits: AI Generated

DOI: 10.1016/j.clet.2026.101327

Keywords: solar PV waste, bi-level optimization, circular economy, reverse logistics, Weibull distribution, vehicle routing problem, Stackelberg game, recycling incentives, extended producer responsibility, Jaipur case study, e-waste management, renewable energy

The core of the framework is a re-optimization subproblem the authors call the adaptive green vehicle routing problem, or AG-VRP. The green vehicle routing problem, originally developed in operations research to model fleets of alternative-fuel vehicles with limited range and mandatory refueling stops, assumes that every vehicle starts and ends at a single depot. That assumption collapses in a replanning scenario: when an agent is lost or a task set changes, the surviving agents are scattered across the mission area with varying amounts of fuel remaining, and they must continue from wherever they happen to be. The AG-VRP therefore allows arbitrary start locations for each agent, heterogeneous initial fuel states, and a single shared destination, while still requiring that every task be visited exactly once and that refueling depots be used as needed to keep each vehicle within its maximum travel range.

To guarantee that this subproblem can be solved to proven optimality when time permits, the researchers formulated it as a mixed-integer linear program. The formulation tracks binary decision variables indicating which arcs each agent traverses, continuous fuel variables that decrease with distance traveled and reset to maximum capacity at depots, and a sequence variable that eliminates subtours, the closed loops disconnected from the start and end nodes that plague vehicle routing formulations. Virtual depots, generated by copying each physical refueling location, allow depots to be visited multiple times by different agents. This mathematical model serves a dual role: it defines the theoretical lower bound against which faster methods are measured, and it provides optimal reference solutions in small problem instances where a commercial solver can exhaust the search space.

Optimality, however, is a luxury that mid-mission replanning cannot afford. The authors set a fixed computation budget of ten seconds per replanning epoch, reflecting the operational reality that a drone hovering with dwindling fuel cannot wait minutes for a solver. Within that budget, exact mixed-integer programming quickly becomes intractable as the number of tasks grows. The team therefore built a hierarchical solution suite. The first layer is a modified sequential greedy algorithm, a fast heuristic that assigns tasks to agents one at a time using a weighted cost that combines the distance to a candidate task with the distance from that task to the destination. The weighting factor discourages agents from picking up tasks near the end node early in their routes, which would otherwise produce inefficient gradual-approach behavior. The greedy algorithm also embeds explicit fuel logic: an agent only accepts a task if its remaining range can cover the trip to the task plus the worst-case distance onward to a depot or the destination, and it falls back to progressively less attractive tasks when the best candidate is fuel-infeasible.

The second and centerpiece layer is a reinforcement learning approach designed to overcome the myopia of greedy assignment. The authors cast the AG-VRP as a Markov decision process in which each decision step selects one agent and one node for it to visit next. The state captures every agent’s current location, remaining fuel, and partial trajectory, along with the set of unserved tasks. The reward is simply the negative of the total distance traveled by all agents, awarded only at the end of an episode, so maximizing expected return is mathematically equivalent to minimizing total travel distance. Training uses the REINFORCE gradient estimator with a rollout baseline, a technique in which a frozen copy of the policy serves as a comparison standard and is updated only when a paired statistical test confirms that the current policy has genuinely improved.

The neural architecture borrows from the attention models that have transformed combinatorial optimization in recent years. An encoder embeds all nodes of the problem, distinguishing depot nodes from task and start nodes and from the end node, and applies multi-head self-attention layers to capture the relationships among them. On the decoding side, the framework introduces two specialized modules: a vehicle selection decoder that chooses which agent should move next, and a node selection decoder that chooses where that agent should go. Crucially, each agent maintains a hidden state in a long short-term memory network that summarizes its entire partial trajectory, including refueling decisions, so the policy can reason about history rather than just the current snapshot. Remaining fuel is concatenated with the trajectory embedding before a linear layer produces the final vehicle representation, allowing the network to weigh range constraints directly when scoring candidate actions.

In deployment, the system does not gamble on a single method. At each replanning epoch, the mission snapshot, comprising agent positions, fuel levels, and remaining tasks, is converted into an AG-VRP instance, and both the greedy algorithm and the trained reinforcement learning policy run independently. The framework, dubbed RL plus MSGA, then selects whichever solution yields the lower total travel distance. Because the two methods have comparable runtimes, this selection strategy improves solution quality at essentially no additional computational cost. The case study illustrates the framework in action: three agents tasked with thirty tasks and supported by two refueling depots lose one of their number after ten tasks are completed, and the two survivors must absorb the remaining workload. The replanned routes produced by the exact solver, the learned policy, and the greedy heuristic are visually similar, but the reinforcement learning method achieves a smaller optimality gap than the greedy baseline.

The numerical experiments scale this evaluation dramatically, covering one thousand problem instances across six task counts from ten to one hundred and two fleet sizes, for a total of twelve thousand problem geometries. The results reveal a striking divide. As problem size grows, the proportion of instances for which the exact solver can even find a feasible solution within ten seconds falls sharply, whereas the proposed heuristics deliver feasible solutions in under one second with a one hundred percent success rate across all scenarios. On instances where the exact solver proves optimality, the reinforcement learning policy consistently lands closer to the optimum than the greedy method, and the combined strategy achieves the smallest gaps overall. Most remarkably, for the largest instances, where the solver struggles to find any high-quality feasible route, the learned policy sometimes matches or beats the solver’s output, producing negative optimality gaps.

Beyond raw distance, the study examined mission-oriented metrics that matter to operators. The reinforcement learning policy consistently leaves the largest fuel reserve at mission completion, indicating a more conservative routing style with a bigger safety margin, while the combined strategy minimizes the number of refueling operations without sacrificing route quality. Workload balance, measured by the standard deviation of tasks assigned to each agent, remained comparable across all methods, showing that the quality gains of the learned policy do not come at the price of overburdening individual vehicles. Even for one-hundred-task instances where optimal solutions could still be found, the combined approach kept optimality gaps below ten percent, a level of performance the authors attribute to the complementary strengths of learned global structure and fast greedy feasibility.

The implications extend beyond military surveillance and reconnaissance, the scenarios that motivate much of this line of research. The same mathematics applies to planetary surface exploration, where in-situ resource utilization stations serve as refueling depots for rovers, and to green logistics fleets of electric or alternative-fuel delivery vehicles that must recharge mid-route. The authors note that future work could incorporate kinematic path constraints such as minimum turning radius, which would replace straight-line distances with Dubins paths and make travel cost dependent on vehicle heading, as well as time windows, agent-specific tasks, and inter-agent collaboration. For now, the study demonstrates that a carefully trained attention-based policy, paired with a fuel-aware greedy fallback, can keep a fuel-constrained drone team flying coherently through the kind of mid-mission shocks that would ground a statically planned fleet.

Subject of Research: Real-time replanning of fuel-constrained multi-UAV missions using reinforcement learning and greedy heuristics

Article Title: Adaptive Planning for Multiple UAVs with In-Flight Refueling

Article References: Gwon, B., Choi, E., Lee, J., & Ahn, J. (2026). Adaptive Planning for Multiple UAVs with In-Flight Refueling. International Journal of Aeronautical and Space Sciences. https://doi.org/10.1007/s42405-026-01249-w

Image Credits: AI Generated

DOI: 10.1007/s42405-026-01249-w

Keywords: UAV, reinforcement learning, vehicle routing problem, in-flight refueling, rolling-horizon replanning, attention model, mixed-integer linear programming, greedy algorithm, LSTM, dynamic environments, fuel constraints, KAIST