Dynamic Planning Research Articles

This research addresses the optimal management of the Zongo II hydroelectric dam reservoir, located on the Inkisi River in the Democratic Republic of Congo. Faced with the persistent energy deficit in the city of Kinshasa, the main objective is to define a sequence of optimal decisions aimed at maximizing energy production during the dry season. The study is based on hydrological and hydraulic modeling of the site, taking into account the topography, geology, and hydrological characteristics of the river. From these elements, a mathematical model of the management system was developed and then optimized through dynamic programming. The results make it possible to formulate rules for the daily management of reservoir levels, ensuring stable turbine operation while reducing water losses. This work illustrates the importance of rational and dynamic planning in the management of hydraulic infrastructures, especially in tropical contexts subject to strong seasonal variability. The developed approach can serve as a reference for integrated water resources management and other hydroelectric projects in the DRC.

The analysis of spatial patterns and variations in urban land prices aids urban planning, socioeconomic analysis, investment decisions, resource allocation, and monitoring of urban spatial growth and land market dynamics. However, scholarly research on the spatial patterns of urban land prices in Sub-Saharan African cities with predominantly informal land markets and settlements remains limited. This study applied spatial statistics to analyse the spatial autocorrelation of residential land prices (RLPs) in Dar es Salaam, Tanzania, aiming to understand their spatial distribution and variation. Global indicators of spatial association (GISA) and Local Indicators of Spatial Association (LISA) were utilised, calculating global Moran’s I and local Moran’s I, respectively, using 452 RLP data from 2020 collected by the Government Chief Valuer. GISA results revealed highly clustered RLPs with strong positive spatial autocorrelation (Moran’s I = 0.83). LISA analysis identified clusters of sub-wards with lower RLPs below the mean, dominating the city's land market. Statistically significant and non-significant LISA results delineated peri-urban and rapidly growing areas. This study provided evidence-based insights for urban planning, policies, infrastructure development, and investor decisions, highlighting the importance of spatial statistics at the regional and sub-regional levels in understanding and improving urban dynamics and land market efficiency.

Abstract This paper studies integrated production and safety stock planning in high-tech low-volume manufacturing supply chains facing various complexities. Multiple end items assembled from a large number of materials, with intricate production processes and component commonality, result in a general-structure multi-echelon network subject to uncertain lead times and capacity constraints. Furthermore, extensive overall lead times necessitate planning based on forecasts of highly uncertain demand, which exhibits non-stationarity due to business cycles. We introduce a data-driven rolling horizon framework that combines a dynamic tactical production-inventory planning model, incorporating demand forecast and supply progress updates, and a strategic safety stock placement heuristic. A distinctive feature of our integer programming-based production planning model is the inclusion of safety stock replenishment decisions following a given safety stock policy, thereby acknowledging the crucial interplay between production and safety stock planning in capacitated systems. Our heuristic effectively exploits the production planning outcomes to derive efficient safety stock policies, trading off inventory investment and customer service. We provide a comparative study with planning practices at our industry partner, ASML. Our findings show that existing hedging, based on safety stocks derived from classical stochastic inventory models, results in service levels 13 percentage points below target due to limiting assumptions regarding demand uncertainty and capacity constraints. Moreover, we highlight the importance of an integrated approach towards multiple uncertainties to benefit from risk pooling, where current separated strategies require a 9% increase in inventory investment when additionally including lead time uncertainty, while our methodology maintains delivery performance with minimal extra cost.

A cumulative capital approach for dynamic transmission expansion planning: enhancing cost efficiency and grid development

The adoption of automated visual inspection systems is growing across various industries, such as manufacturing and energy, and is expected to expand significantly into other sectors, including aerospace. However, these systems often encounter challenges when visually inspecting highly reflective metallic surfaces, as varying light conditions can obscure critical surface details, thus risking errors in defect detection. This paper addresses these challenges by developing methods to detect specular light reflections in order to automatically assess the quality of inspection images. This enables automated systems to take inspection images from different angles, to avoid undesired reflections. We show that U-Net based architectures trained on a novel dataset of inspection images and reflection masks lead to good detection under challenging conditions. The results demonstrate that Convolutional Neural Network (CNN)-based models, particularly U-Net++ with a ResNet-50 encoder, outperform Transformer-based approaches, achieving the highest accuracy in identifying reflective areas. While the proposed UNETR-Attention Fusion (UNETR-AF) model shows promise for smaller reflections, it struggles with larger ones. This research offers a practical solution for industries aiming to improve visual inspection reliability, particularly for safety-critical applications. By enabling automated systems to handle reflective surfaces effectively, it addresses a significant gap in current inspection technologies.

GeoPF: Infusing Geometry Into Potential Fields for Reactive Planning in Non-Trivial Environments

Next-generation artificial intelligence (AI) ecosystems are redefining the landscape of life science research by enabling real-time biosurveillance, precision health analytics, and dynamic intervention planning. The convergence of high-throughput biological data, advanced computational architectures, and intelligent analytics has created an unprecedented opportunity to transition from reactive to predictive and preventive healthcare models. Through continuous data assimilation from genomic, proteomic, and clinical sources, AI systems can detect early biomarkers of disease outbreaks, assess evolving population health risks, and support precision-guided treatment interventions. These frameworks not only enhance early detection of infectious agents and genetic predispositions but also streamline response coordination across global health networks. At the core of this transformation lies the integration of multi-agent AI platforms, federated learning pipelines, and graph-based causal reasoning that enables adaptive modeling of complex biological interactions. Predictive biosurveillance tools, powered by these AI ecosystems, can identify anomaly patterns in pathogen evolution, track transmission networks, and forecast therapeutic efficacy with high temporal precision. Such innovations are essential for developing resilient public health infrastructures capable of responding to pandemics, antimicrobial resistance, and chronic disease management. I will execute a focused agenda to deliver clinically reliable, secure, and compliant healthcare AI with explicit milestones and KPIs, including (1) AI-driven biosurveillance systems integrated with secure data streams, (2) precision health analytics platforms for individualized risk prediction, and (3) dynamic intervention engines governed by ethical and transparent decision protocols. Collectively, this initiative will accelerate scientific discovery, strengthen biosecurity, and establish an intelligent foundation for global life science innovation.

Path planning is essential for mobile robots to operate efficiently in uncertain environments. As the number of robots increases, centralised approaches struggle to provide feasible solutions within real-time constraints. To address this, an adaptive hybrid strategy-based decentralised path planning algorithm is proposed to solve the real-time path planning problem through decentralised computing. First, various path planning strategies are introduced, including a neural computing planning strategy, a dynamic search planning strategy, and a cluster coordination planning strategy. Then, an intelligent strategy selection mechanism with an adaptive strategy adjustment factor is designed, allowing each robot to dynamically select optimal planning strategies based on their current state and ensuring the planned paths exhibit greater flexibility and adaptability. Finally, the results of the ablation experiment indicate that all three strategies effectively enhance the navigation capabilities of robots in a decentralised mode. The algorithm comparison experiment demonstrates that the proposed algorithm achieves a higher task completion rate and a lower detour percentage in various environments. The decision response experiment shows that our approach has an average decision-making time of approximately 60 ms, which meets the real-time requirements for decentralised path planning of mobile robots in most scenarios.

Effective path planning is crucial for autonomous underwater vehicles (AUVs) in rescue and logistics operations. This paper proposes a Risk Aware Proximal Safe Deep Q-Network (RAPS-DQN) to address dynamic path planning challenges across complex underwater terrain. Traditional deep reinforcement learning (DRL) methods show limited generalization in high-dimensional environments, making the adaptive RAPS-DQN approach essential. The proposed method enhances the original DQN by incorporating Lyapunov stability criteria and selectively prioritizing significant transitions based on temporal-difference errors. The suggested approach develops paths that satisfy requirements both efficiently and steadily, as shown by real-time testing and comparison with traditional DRL techniques. The findings show RAPS-DQN facilitates effective learning in complex AUV navigation scenarios, with potential applications in underwater search and rescue missions in flooded urban environments and disaster-affected areas.

Traffic flow prediction is an essential part of intelligent transportation systems (ITS), facilitating dynamic traffic control, congestion alleviation, and route planning. The past few years have seen the booming of Graph Neural Networks (GNNs), providing a strong tool to capture the sophisticated spatial compositions and dynamic temporal patterns inherent in urban road networks. In this paper, I provide a systematic and comprehensive review of GNN-based solutions for short-term traffic prediction. I first briefly review the basic concepts and categorization of GNNs, then elaborate on representative models such as spatial and spectral convolutional networks, temporal graph structures, and hybrid models with attention mechanisms and recurrent units. Besides, I recap typical traffic datasets and evaluation metrics, compare the performance of different models from multiple aspects, and highlight key technical challenges such as spatiotemporal heterogeneity, scalability, and interpretability. Finally, I suggest future research lines to improve the accuracy, efficiency, and robustness of GNN-based traffic prediction models for real-world ITS applications.

To investigate the role of an Internet-Plus-Based smart emergency platform in reducing pre-hospital Medical Priority Dispatch System (MPDS) response time and optimizing dynamic allocation of emergency resources, providing evidence-based support for enhancing emergency care efficiency, a mixed approach was employed including: (1) Quantitative analysis: Comparing response time and dispatch efficiency data (n=12,358 cases) from six months before and after the launch of an Internet-based emergency platform in a city (July 2024–June 2025); (2) Qualitative research: Conducting semi-structured in-depth interviews with 30 pre-hospital emergency nurses and 10 dispatchers, using Colaizzi's phenomenological analysis method to extract themes and analyze platform application pain points and improvement directions. The study showed that following platform implementation: Average dispatch response time (from call receipt to vehicle dispatch) decreased from (92.5 ± 15.8) seconds to (38.2 ± 9.4) seconds (t=15.324, P<0.001); and average dispatch response time (from assignment to departure) decreased from (135.6 ± 20.1) seconds to (98.7 ± 14.5) seconds (t=8.912, P<0.001) dramatically in respective. Meanwhile, resource allocation: The proportion of cross-regional collaborative ambulance assignments increased from 15.7% to 28.9% (χ²=210.5, P<0.001). The “resource misallocation rate” (e.g., dispatching non-critical cases to critical care units) based on platform AI triage decreased from 12.5% to 5.8% (χ²=95.7, P<0.001). Besides, nurse satisfaction with “intelligent triage guidance” and “dynamic route planning” functions reached 92%. In conclusion, in the era of Internet Plus and AI, the smart emergency platform integrates data and enables intelligent decision-making, significantly optimizing pre-hospital response workflows and resource allocation efficiency. This represents a core implementation pathway for “Internet Plus Emergency Nursing” and holds significant implications for improving pre-hospital emergency response time. Therefore, future efforts should be focused on balancing technological empowerment with humanistic care while strengthening nurses' information literacy training.

Advancing autonomous driving technology demands balancing multiple evaluation metrics, including ride comfort, energy efficiency, and vehicle performance. This study presents a novel steering control system that incorporates the 6-DoF-SVC model into the reward function of a DDPG framework, enabling the optimization of ride comfort while preserving energy efficiency and vehicle performance. The proposed system utilizes a hybrid architecture combining DRL-based decision-making with a PI controller, allowing flexible trade-offs among evaluation metrics. Numerical experiments conducted in MATLAB and Simulink under varying ride comfort thresholds (thsvc) demonstrate that appropriate thsvc settings can simultaneously enhance ride comfort and either energy efficiency or vehicle performance, depending on operational objectives. Additionally, the study identifies a significant trade-off between steering effort and ride comfort, indicating that specific thsvc settings permit the simultaneous optimization of both. However, challenges such as oscillatory behavior during lane changes were observed, highlighting potential areas for improvement. This research provides valuable insights for designing autonomous vehicle control systems that address competing objectives. Future work will focus on dynamic velocity planning, integrating more realistic driving scenarios, and validating system performance in dynamic environments involving other vehicles and pedestrians. By addressing these challenges, the proposed system aims to improve the safety, efficiency, and comfort of autonomous driving solutions.

This paper explores trajectory planning for robotic arms in 3D operational spaces. To improve the adaptation to dynamic via‐points in complex, confined environments, an enhanced dynamic movement primitives (DMP) approach is proposed and designed for dynamic planning under composite steering force field constraints. By incorporating steering attraction forces, this method enhances the generalization capability of DMP, allowing the robotic arm to navigate through dynamic via‐points flexibly without altering the start and end positions. The trajectory shape is adjusted via regression attraction forces, which helps preserve the demonstrated trajectory, reduce free‐space loss, and improve the system's adaptability to complex, dynamic environments. The convergence of the target state is rigorously proven using Lyapunov stability theory. Numerical simulations and experiments conducted with the Franka robotic arm validate the effectiveness of the proposed approach. Results show that in dynamic environments with multiple via‐points, this method produces reliable trajectories for robotic arm movements, significantly enhancing the adaptation of DMP to dynamic contexts. The planning process requires no additional learning, and the generated trajectory closely resembles the original demonstrated path. This method enables effective via‐point operations in confined spaces without requiring additional learning, while maintaining existing skills.

Iguaçu National Park (INP) receives approximately 2 million visitors per year, which means not only a huge management challenge but also offers great opportunities for communicating with this audience. Despite this significant amount, the park never had an environmental interpretation plan. The Iguaçu National Park Environmental Interpretation Program was designed to improve the visitor experience through a more effective communication with the different subgroups of visitors, prior, during, and after the visitation experience to the PNI. Documentary analyses were carried out considering the management plan, the public use plan, thematic publications etc., and also nine remote (virtual) meetings were held with specific groups of key stakeholders, totaling more than 70 people involved, which provided support for the development of the interpretation program. Six target audience groups considered to be priorities were identified and eighteen interpretative goals were established covering three dimensions: cognitive, emotional, and behavioral. A main theme for interpretation of the park and four sub-themes were developed, each addressing a group of resources collectively identified. Finally, minimum guidelines were produced for twenty-seven interpretative projects, each one establishing the target audience, application/implementation location, interpretative goals and themes to be worked on, in addition to the expected deadline, among other guidelines. Until then, the experiences of gathering input for environmental interpretation planning within the scope of ICMBio had been through in-person workshops that brought together the diversity of key stakeholders involved in the management of the protected area at the same time. The holding of specific and online workshops proved to be satisfactory, being fundamental for the construction of the Program. Considering that the document is part of a dynamic and adaptable plan, which must be integrated into a PDCA management model, it is expected that, for the next revisions of the Environmental Interpretation Program, there will be monitoring of the implemented interpretative projects, as well as mapping of the profiles of the INP visitors. Such monitoring data will make it possible to identify the effectiveness of the projects and the more specific perceptions, demands and preferences of each niche, allowing for the adaptation and improvement of the interpretation strategies.

This article addresses the long-term profit maximization needs of a rural village in the mountainous region of North China under limited arable land resources, focusing on the optimization of crop planting across multiple cycles. The study particularly tackles the challenge of coordinating yield overproduction risks with complex constraints. Existing research predominantly focuses on single-year or single-objective scenarios, lacking systematic integration of long-term dynamic planning and differentiated sales strategies. Our research constructs two profit functions based on a linear programming model, incorporating a seven-year cycle, land type restrictions, and crop rotation constraints to design a phased optimization strategy. By integrating land parcel information, crop attributes, and historical planting data, the impact of overproduction scenarios on profits is quantified. Under the assumption of a static market, the price reduction sales strategy outperforms the stockpiling strategy, increasing total profits by 12.5%-20%. Additionally, crop rotation constraints balance short-term gains with long-term soil health, and post-optimization, high-value crops such as golden oyster mushrooms dominate the planting area, validating the model's adaptability to complex constraints. This experiment proposes a planting optimization scheme that combines economic efficiency with sustainability by integrating multi-cycle dynamic planning and differentiated sales strategies, providing theoretical support for rural agricultural decision-making. Future work could enhance the model's dynamism by incorporating climate forecasting and blockchain technology.

The adoption of artificial intelligence has risen, yet research on its impact on innovation processes between actual businesses remains sparse. This research fills the present gap by investigating ten workers from a tech startup who utilize artificial intelligence tools in operational and creative activities. The paper analyzes business-related AI functionality through a qualitative analysis of ten tech start-up employees. The examination reveals that AI produces significant enhancements in problem resolution by executing mundane actions while analyzing large datasets to deliver data-driven suggestions to users. The interview respondents mentioned that AI’s role in diminishing supply chains is 15%, while allowing AI to manage customer service without employee engagement in 80% of interactions. The implementation costs, along with data dependency and occasional contextual blindness in AI systems, represented some of the problems in this system. Analysis demonstrated that AI tools enable the development of innovative concepts and challenge established viewpoints, prompting participants to create a gamified loyalty system and dynamic content planning. Participants in the study emphasized the need for human involvement to refine AI-based insights, recognizing how human imagination complements AI capabilities effectively. The work enhances academic discussions about AI-related problem-solving and creativity while offering specific business-related recommendations for implementation. The recommendations begin with establishing initial experimental programs, while providing support for employee’s skills development, and fostering strong alliances between technical AI personnel and professional subject matter experts. Research topics focused on AI application fields and the anticipated impacts on company decision-making, as well as the ethical ramifications, need further exploration. This research confirms the revolutionary potential of artificial intelligence systems for problem-solving methods, but requires proper execution, along with human supervision, to fully realize their advantages.

Rescue transportation will be severely hindered under urban waterlogging disasters. To solve the problem of delayed rescue of disaster vehicles, this study is based on a hydrodynamic model and uses a 2D hydrodynamic model to numerically analyze urban waterlogging. At the same time, a 1D pipeline network is used to calculate pipeline hydrodynamics and urban hydrodynamics modeling is completed by coupling 2D hydrodynamics with a 1D pipeline network. Considering the characteristics of urban transportation, static and dynamic path planning techniques are introduced to complete vehicle rescue path planning. In the analysis of hydrodynamic models, the monitoring value of the research model at the B water accumulation location was 41.2 cm, which was closer to the actual value and had a lower relative error, outperforming similar models. In addition, in the analysis of multiple rainfall scenarios, the proportion of short-term waterlogging caused by high-intensity rainfall was relatively high. For example, in scenario 4, the proportion of waterlogging formation within half an hour was 8.8%, which was higher than that of low rainfall. In addition, in static path planning, the research technique took 821s in scenario 4 and had a shorter planning distance, which is superior to similar techniques. In dynamic programming, the research technique performed better overall with a driving time of 1,054s in scenario 5 and a shorter planning distance of 7,723m. The research technology has good application effects in urban waterlogging disaster rescue. This research will provide technical support for urban disaster analysis and rescue.

The accelerating pace of urbanization has significantly complicated traffic management systems, leading to mounting challenges, such as persistent congestion, increased travel delays, and heightened environmental impacts. In response to these challenges, this study presents a novel deep learning framework designed to enhance short-term traffic flow prediction and support intelligent transportation systems within the context of smart cities. The proposed model integrates Gated Recurrent Units (GRUs) and Long Short-Term Memory (LSTM) networks, augmented by an attention mechanism that dynamically emphasizes relevant temporal patterns. The model was rigorously evaluated using the publicly available datasets and demonstrated substantial improvements over current state-of-the-art methods. Specifically, the proposed framework achieves a 3.75% reduction in the Mean Absolute Error (MAE), a 2.00% reduction in the Root Mean Squared Error (RMSE), and a 4.17% reduction in the Mean Absolute Percentage Error (MAPE) compared to the baseline models. The enhanced predictive accuracy and computational efficiency offer significant benefits for intelligent traffic control, dynamic route planning, and proactive congestion management, thereby contributing to the development of more sustainable and efficient urban mobility systems.

A novel impact-based dynamic motion planning of parallel kinematic forming robot under heavy load

Dynamic preference-based multi-modal trip planning of public transport and shared mobility