Matteo Salani

In this work, we investigate whether machine learning can be leveraged to identify promising states in dynamic programming algorithms, focusing on Elementary Resource Constrained Shortest Path Problems (ERCSPP). More in detail, we solved 41 single resource instances from SPPRCLIB using iterative relaxation techniques through the PathWyse library, systematically collecting all generated states (i.e. labels). We designed ad-hoc features computable in constant time and constructed two datasets: one containing all generated labels (G) and another with only those inserted into data pools (I), totaling several hundred million labels. Machine learning tools are then employed to explore these datasets, revealing significant patterns between successive relaxations. Leveraging these insights, we propose a normalization approach and apply supervised learning techniques to distinguish dominating states, both within subsequent relaxations of the same problem and in previously unseen instances. Our results demonstrate the effectiveness of this approach on Dataset G, while for Dataset I, performance varies, showing strong results within the same instance but declining for unseen ones. We therefore propose a data-driven policy that leverages trained models to prioritize the extension of promising labels during dynamic programming. Preliminary evaluation shows consistent reductions in the number of extended states across almost all instances, with speedups emerging on larger ones, supporting the viability of data-driven dynamic programming algorithms.

The most successful approaches for solving the Elementary Resource Constrained ShortestPath Problem (ERCSPP) to optimality are based on Dynamic Programming (DP) combinedwith an iterative relaxation strategy. Decremental State Space Relaxation (DSSR) and ng-path relaxation have long been established in the literature as relaxation techniques.In DP algorithms, an effective policy for selecting the next state to extend among the openones is crucial for performance. In a first exploratory work, we showed that machine learningcan be used to identify promising states in DP, Basso and Salani (2026). In particular, if aclassifier is trained to identify dominated states in a given DSSR iteration, it can then beused to estimate the quality of the states generated in the subsequent iteration.Building on this idea, in this work we show that it is possible to efficiently embed such amodel into a state-of-the-art DP algorithm for the ERCSPP and use its predictions to definea data-driven state exploration policy. Although the approach is data-driven, it remainsexact and does not require pre-existing training data. The algorithm has been integratedinto the Pathwyse library and tested on a dataset of multi-resource instances. Comparedwith a node-based extension policy, the proposed approach consistently reduces the numberof explored DP states and, for sufficiently difficult instances, also improves runtime.

Multi-agent systems (MAS) based on large language models (LLMs) are on the rise due to their improved capabilities compared to single agents. In particular, among the emerging MAS for software development, those establishing automated agentic workflows contribute to the idea that software development is an essentially solved task. This study aims at critically assessing five MAS for software development that cover multiple phases and support automated workflow: MetaGPT, ChatDev, AgileCoder, HyperAgent and AgentLite. Here, 14 research questions are identified, specifically referring to architectural-, technical-, operational-, human interaction- and responsible AI-aspects. The findings from the study can help interested researchers and practitioners to understand which design choices bring advantages for particular aspects and where, on the contrary, existing solutions are suboptimal and what could be suggestions for addressing the gaps. Due to its research question-oriented layout, this paper is ideal as a reference which, for a particular aspect, facilitates grasping the overview of main design choices, capabilities, limitation and reconciliation approaches.

Small hydropower plants (HPPs) equipped with water storage play an important role in managing fluctuating energy demand. This article presents a real-world case study in which model predictive control (MPC), driven by energy-demand and water-inflow forecasts produced using the Light Gradient Boosting Machine (LGBM), is applied to optimize the operation of a small hydropower plant for peak shaving. A comparative analysis is conducted between the current non-predictive control strategy, which relies on operator decisions for peak shaving, and a fully automatic controller that optimally schedules the utilization of available water resources based on ML predictions. Results show that the MPC can outperform the operator-based scheduling and that this has the potential to improve the peak shaving capabilities of small HPPs. Unlike previous studies that predominantly focus on large and complex hydropower systems or introduce new control formulations evaluated under idealized assumptions, this work offers a pragmatic solution to the underexplored context of peak shaving for small HPPs operated with limited data and resources, that small utilities can adopt with minimal effort using their own data. We show that even these small-scale hydropower operations have room for improvement through optimal scheduling.

Small hydro power plants (HPPs) play an important role in managing fluctuating energy requirements. This article presents a real-world case study where model predictive control (MPC) utilizing lightGBM-based machine learning (ML) forecasts of energy demand and water availability is employed to optimize the scheduling of a small HPP for peak shaving. A comparative analysis is conducted between the current non-predictive control strategy, which relies on operator decisions for peak shaving, and a fully automatic controller that optimally schedules the utilization of available water resources based on ML predictions. Preliminary results show that the MPC can outperform the operator’s decisions and that this has the potential of improving peak shaving capabilities of small HPPs, emphasizing the role of predictive control methodologies for exploiting energy storage resource in the management of the distribution grid. This approach offers a pragmatic solution that small utilities can adopt with minimal effort using their own data.

Aligned with the objectives of the energy transition, the increased penetration levels of electric vehicles as part of the electrification of economy, especially within the framework of local energy communities and distributed energy resources, are crucial in shaping sustainable and decentralized energy systems. This work aims to assess the impact of escalating electric vehicles’ deployment on sustainable local energy community-based low-voltage distribution networks. Through comparative analyses across various levels of electric vehicle integration, employing different charging strategies and system management approaches, the research highlights the critical role of active system management instruments such as smart grid monitoring and active network management tools, which significantly enhance the proactive management capabilities of distribution system operators. The findings demonstrate that increased electric vehicle penetration rates intensify load violations, which strategic electric vehicle charging management can significantly mitigate, underscoring the necessity of load management strategies in alleviating grid stress in the context assessed. This study highlights the enhanced outcomes derived from active system management strategies which foster collaboration among distribution system operators, demand aggregators, and local energy communities’ managers within a local flexibility market framework. The results of the analysis illustrate that this proactive and cooperative approach boosts system flexibility and effectively averts severe grid events, which otherwise would likely occur. The findings reveal the need for an evolution towards more predictive and proactive system management in electricity distribution, emphasizing the significant benefits of fostering robust partnerships among actors to ensure grid stability amid rising electric vehicle integration.