Biomass supply chains face both internal challenges (low energy density, bulkiness, seasonality) and external challenges (market uncertainty). We developed a two-stage stochastic mixed-integer linear programming (MILP) model that configures an (s, S) inventory control policy for a willow-based pellet production facility. This model accounts for seasonal willow harvesting capacity and stochastic daily wood pellet demand. Using a case study of a proposed wood pellet production facility in Schenectady County, New York, USA, we evaluated model-suggested inventory control policies using multiple out-of-sample pellet demand scenarios by assuming a moderate demand uncertainty with a 10% coefficient of variation. Inventory policies acquired from modeling stochastic demands resulted in a 1% mean cost reduction compared to the deterministic approach. More importantly, these policies reduced cost variability by 44% under moderate demand uncertainty. We expected the performance advantage of the stochastic model to increase as uncertainty levels rise. Test cases show that frequent inventory policy adjustments provide additional cost savings. The model successfully accounted for seasonal supply constraints and stochastic market demand to facilitate a multi-feedstock strategy that offers additional supply chain resilience and associated cost reduction. Overall, the stochastic modeling framework provides facility managers with more robust inventory planning under real-world constraints. The framework’s computational efficiency and broad applicability make it suitable for adoption by diverse biomass industries with uncertainties in their supply chains, particularly those facing seasonal feedstock supply constraints and stochastic end product demands.
Forest road surfaces are subject to continuous deterioration from environmental factors, even in the absence of regular traffic, creating significant challenges for safety and maintenance planning. Traditional manual inspections are labor-intensive, costly, and often impractical for extensive networks. This study introduces an Artificial Intelligence (AI)-powered framework, leveraging advanced computer vision techniques, to automate the visual inspection of forest road surfaces using high-resolution UAV imagery. We evaluated three state-of-the-art deep learning models, YOLOv5, YOLOv8, and Faster R-CNN, on a custom dataset of potholes, rutting, and stone protrusions from a mountainous forest road in Iran. The YOLOv5 model demonstrated superior performance, achieving a mean Average Precision (mAP@0.50) of 0.56 and an overall recall of 0.459. It proved highly effective at detecting common distresses like rutting and potholes, while also offering the most robust performance for challenging stone protrusions. These findings validate the integration of AI and computer vision as a transformative tool, providing forest engineers with a scalable, objective, and data-driven methodology to enhance safety, optimize maintenance budgets, and support sustainable forest management.
Sustainable forest management relies on effective operational planning to ensure that harvesting practices support long-term objectives. Operations research methods have largely been used to support operational decision-making in forest harvest planning but the broader strengths, limitations and barriers to adoption remain unclear. This review addresses this gap by synthesizing existing research on operational planning tools for forest harvesting. Using PRISMA protocols, we conducted systematic searches in Scopus and Web of Science and identified 23 peer-reviewed studies published between 2005 and 2024. The included studies employed diverse approaches across geographic regions, most commonly being mixed-integer programming and geographic information systems (GIS). Results show that while these models provide valuable insights and demonstrate technical expertise, they are often hard-coded to specific sites, lack reproducibility and are rarely open-source. Developing modular, transparent and user-centric tools could strengthen the existing connection between research and practice, enabling forest planners to manage uncertainty and improve efficiency while aligning with broader sustainability goals. Our findings highlight the importance of designing adaptable frameworks that embed site-specificity as a structural element rather than a limitation. We synthesize findings into a practitioner checklist, covering inputs, constraints, solution approach, validation, user experience and openness to guide tool design and evaluation.
In ecosystem-based forest management, forest roads are expected to support ecological integrity alongside economic, social, and technical functions. However, conventional cost-oriented evaluations often fail to capture this multidimensional performance. This study evaluates the functional efficiency of forest roads using a Data Envelopment Analysis (DEA) framework integrated with Geographic Information Systems (GIS). This study aims to develop a multidimensional efficiency assessment framework to support forest road planning while minimizing adverse impacts on sensitive natural areas, including biodiversity corridors, water resources, and protected forest zones. Four functional scenarios were considered: ecological, economic, social, and technical. Thirty forest roads located in the Ma & ccedil;ka Forest Sub-District Directorates (T & uuml;rkiye) were analyzed as Decision Making Units (DMUs) using an input-oriented CCR DEA model. Scenario-specific input and output variables were derived from GIS-based spatial analyses and field measurements. Twenty-one variables were quantified via GIS models and eighteen variables via field surveys, reflecting terrain, geometry, environmental sensitivity, and accessibility. The results reveal clear functional contrasts among road segments. In the ecological scenario, 24 out of 30 roads (80.0%) were classified as efficient, followed by the economic scenario with 22 efficient roads (73.3%). Technical and social scenarios exhibited lower proportions of efficient roads, with 17 (56.7%) and 14 (46.7%) efficient roads, respectively. Ecologically efficient roads were associated with stable terrain, lower landslide occurrence, and stronger forest connectivity. Economically efficient roads showed appropriate spacing and limited surface deformation, while technical efficiency was mainly linked to compliance with geometric standards. This DEA-GIS framework supports function-oriented decision making in forest road planning.
Several models exist for estimating energy consumption in both diesel and electric trucks. However, there is limited research on hybrid trucks, which combine diesel and electric powertrains. We propose three energy-consumption models to estimate energy use in hybrid-electric trucks. The first model is a standard energy model grounded in fundamental physical principles. The second model estimates the overall energy consumption as a function of road gradient and speed. The third model provides a comprehensive analytical representation of energy distribution in hybrid trucks by separately modeling the diesel engine and electric motor, enhancing its adaptability to diverse driving conditions. All three models account for variations in road slope and speed limits. The models are trained on measured data from energy-consumption tests conducted on a prototype hybrid-electric truck developed by a Canadian research and development organization. The models are then used in a case study to estimate the hybrid truck's performance on roads with varying characteristics.
As biomass energy becomes a key part of the renewable energy mix, optimizing biomass harvesting operations is crucial for ensuring efficiency, economic viability, and a sustainable energy supply. This study analyzed two biomass harvesting operations in the Coastal Plain of Florida, each employing rubber-tired feller-bunchers, grapple skidders, trailer-mounted knuckleboom loaders, and whole-tree woodchippers. While both operations produced wood chips, Operation A harvested material from a stand dominated by small-diameter trees, whereas Operation B harvested understory and unmerchantable trees from a mixed stand where merchantable stems were handled separately. The average productivity rates per productive machine hour (PMH) were 17 green tonnes for the feller-bunchers, 38 green tonnes for the skidders, and 64 green tonnes for the chippers. Low productivity of the feller-bunchers was attributed to the small diameter of harvested trees, with over 60% having a diameter at breast height (DBH) of 7.6 cm or less. Despite chippers having a low utilization rate (58.5%), feller-buncher capacity ultimately constrained overall system productivity. Chipping was the most expensive activity at $USD 105 per scheduled machine hour (SMH), followed by felling ($USD 102), skidding ($USD 98), and loading ($USD 92). Compared with the 2020 TimberMart-South Coastal Plain final-harvest cut-and-load reference rate of $USD 12.5 per tonne, biomass cut-and-load rates were $USD 2.3 higher in Operation A and USD $5.1 per tonne higher in Operation B, with lower costs in Operation A due to higher productivity. This study provides valuable insights into improving the economic efficiency and operational productivity of biomass harvesting systems while supporting the sustainable development of biomass energy.
A functional and low-impact forest road network is essential for sustainable forest management, yet maintaining such infrastructure is costly and requires monitoring tools that are reliable and simple enough for operational use. We present an automated approach to detect, map, and evaluate forest road surface deterioration, designed to support end-users, including those with limited road expertise, to indicate required maintenance actions. The system relies on data collected by the vehicle-mounted near-field sensor platform RoadSens, which integrates stereo camera imagery with GNSS-based geo-referencing to capture detailed road surface information. Collected data are processed within a monitoring and scheduling environment using a YOLOv8 object detection model trained on nearly 14,000 annotated images. The model identifies six key deterioration features: potholes, wheel ruts, gullies, washboards, stones, and vegetation. These detections are used to locate maintenance-relevant features and classify road segments into three deterioration levels based on coverage thresholds, which are then visualized through a traffic-light system. A case study on a forest road in southern Norway demonstrated the system's ability to detect and classify maintenance needs. While performance was strong for more uniform features such as vegetation, irregular structures like wheel ruts proved more challenging, occasionally leading to misclassification of actual maintenance requirements. Nevertheless, the findings confirm the technical feasibility of integrating object detection models into data-driven forest road maintenance scheduling. Future improvements will require larger and more diverse training datasets, as well as classification frameworks tailored to local conditions and specific road-user needs.
Collection, processing and provision of comprehensive geometric information of forest roads is decisive for its technical classification to facilitate sustainable timber supply chains. An automized classification system based on the mobile proximal sensor platform RoadSens was developed, applied and validated through a case study approach in Eastern Norway. Six sample roads of various vegetation stages were surveyed through RoadSens and complemented through sampled total station measurements for validation purposes. The determined geometric parameters road slope, curvature and width were used for technical classification following the national forest road standard. Road width was identified as the main constraint in meeting the standard, resulting in a general downgrading of the sampled roads according to its technical class. The results showed a root mean square error (RMSE) ranging from +/- 0.53 to 1.50 m (12-33%) depending on the road and vegetation stage compared to the validation data. Despite these accuracy constraints, the application case study already indicates a general need for improvement of road data acquisition and updating of associated databases. The study underscores that, despite the challenges and limitations, there is a clear need for an automated sensing and classification system, which offers a cost-effective alternative to manual surveying and requires less specialized expertise.
Forestry operations planning has typically relied on forest inventories, which require manual field measurements, a process that is often costly in terms of money and labor. Advancements in remote sensing technologies have demonstrated the potential to enhance both the accuracy and efficiency of forestry operations planning, thereby reducing costs and labor requirements. Integrating these technologies with forestry machinery necessitates automation, particularly in harvesters. The need for an exact three-dimensional description of tree stems has become a critical requirement. However, stem detection and segmentation from LiDAR data have typically required extensive post-processing, which limits their applicability in scenarios where real-time information is critical. This work examines the feasibility of detecting and segmenting tree stems in real-time using LiDAR data obtained through mobile laser scanning (MLS), creating a proof of concept. To achieve this, each frame captured by the sensor is processed independently and transformed into a 2D projection, where a Convolutional Neural Network (CNN) identifies and segments tree stems. This approach eliminates the need for extensive point cloud analysis, enabling faster processing and rapid response in operational environments. Detecting stems through this method offers distinct advantages over point-based methods. To fully validate this proof of concept, further development and integration of the algorithms will be necessary, along with expanding the training dataset to improve model performance. Accurate real-time detection and segmentation of tree stems and logs could be the basis for the automation and robotization of forestry operations, enhancing efficiency and precision in forest management.
As harvester heads are primarily designed to fell and process uniform conifer stem forms, they struggle with the complex morphology of tropical hardwoods like Acacia mangium, causing significant but unquantified operational inefficiencies. This study aimed to quantify the specific impact of tree characteristics and operational elements on mechanized harvesting efficiency from a work elemental perspective. A continuous time-motion study was conducted on a tracked harvester in a 7-year-old Acacia mangium plantation in Indonesia. Video analysis of 285 uninterrupted productive cycles was performed to extract elemental time consumption, utilizing statistical models and non-parametric tests to evaluate the effects of DBH and stem morphology. DBH was the primary determinant of productivity, yielding an estimated gain of 0.39 ${m<^>{ m{3}}}{ m{/}}PM{H_0}$m3/PMH0 per 1 cm increment. Conversely, stem processing (delimbing and bucking) constituted the critical bottleneck, consuming 45.15% of productive time. Trunk curvature exceeding 15 degrees significantly reduced processing efficiency by 49.8%. Furthermore, boom oscillation damping accounted for 34% of boom extension and grabbing time due to mechanical inertia. Stem irregularity, rather than dimension, is the primary biomechanical constraint limiting harvester performance. Optimizing operations requires advancing harvester head adaptability and boom-control technologies, alongside implementing silvicultural strategies to proactively improve stem straightness and stand uniformity.
Time and motion studies in forest operations benefit from video-based analysis, but manual annotation is time consuming. This pilot study aims to reduce analysis time by developing a deep-learning framework that classifies dashcam video into four work elements: crane out, cutting and processing, driving, and processing. Using a 3D ResNet-50 (PyTorchVideo) trained on manually annotated clips, the model achieved validation F1 = 0.88 and precision = 0.90, showing that spatiotemporal CNNs can capture rele-vant motion and appearance cues in forest environments. Overfitting indicates that more diverse data and better class balance are needed, but the approach shows clear potential to scale automated work-element monitoring and efficiency analysis.
Integrating forest biomass into bioenergy systems poses logistical challenges due to seasonal variations in quality and the dispersed nature of supply. We develop a mixed-integer linear programming model that jointly optimizes procurement timing, multimodal transport (truck-rail-barge), chipping and drying locations, and inventory levels at supply nodes, terminals, and the biorefinery. The model embeds process-state transitions, seasonal moisture profiles, and infrastructure limits. In a large-scale Quebec case study (500 - 3000 dry metric tonne (DMT)/day), integrating rail reduces total system costs by 2.8 - 4.8% and yields mill-gate costs around CAD 119 - 121 per DMT. Terminals near the biorefinery decouple procurement from conversion and support buffer-based strategies through high-moisture periods. The optimization model is computationally tractable and provides a reusable template for planning forest biomass logistics that accounts for seasonal quality, preprocessing, and mode-choice interactions.
Concerns have grown over the environmental impacts of open pile burning, which has the potential to degrade air quality and harm soil. Additionally, pile burning is constrained by narrow burn windows in areas with high wildfire risk. As a result, technologies such as the CharBoss - which converts slash into biochar - are gaining attention. However, optimal deployment remains uncertain because of variability in performance due to pile locations, slash-pile size, and site conditions. We address this challenge with a two-stage simulation - optimization framework. First, a mixed-integer programming model integrated with Monte Carlo simulation selects the cost-minimizing set of deployment sites under uncertainty regarding site conditions and their impact on CharBoss throughput. Second, a capacitated vehicle-routing problem model sequences visits to those sites for fleets deploying multiple units at the same time from a centralized depot. Generally, the model favors piles with large slash volumes on flat terrain near the depot. However, in optimization, some relatively larger piles were skipped because of the longer setup times due to site conditions and long distance to travel to the site. Although direct biochar production constitutes the dominant cost (95-96%), the framework shows that total system cost is minimized by strategically forming larger, accessible piles at sites with low preparation requirements and then sequencing visits efficiently. Our framework was tested with both theoretical and real-world cases, demonstrating its utility in establishing slash-to-biochar strategies as an alternative to traditional pile burning.
Partial and clearcut harvests are common in the Appalachian Mountains, yet few studies have compared harvest area, Best Management Practice (BMP) implementation, and erosion rates of these harvests. We compared 14 different harvest sites looking at harvest area, erosion rates and BMP implementation audit evaluations conducted by the Virginia Department of Forestry. Areas, BMP implementation, and erosion rates were characterized within road, landing, skid trail, stream crossing, harvest area, and streamside management zone operational features. Clearcuts and partial harvests represented 42.9% (n = 6) and 57.1% (n = 8) of harvests respectively. Overall, mountain clearcuts and partial harvests had similar BMP implementation rates (83% and 84%). Erosion rates did not differ significantly between partial harvests (2.35 t/ha/y) and clearcuts (1.50 t/ha/y). Notably, but not significantly, partial harvests had more of the harvest area (7.45%) in skid trails than clearcuts (4.03%). An increase in skid trail area was associated with higher overall erosion rates. Results indicate that while either harvest type can be conducted in a manner that minimizes erosion, increased planning and implementation of additional BMPs, particularly to reduce skid trail area, would be beneficial for both harvest types.
Hurricanes are increasingly frequent and intense disturbances that cause substantial economic and ecological damage to forested landscapes worldwide. Following such events, multiple stakeholders - including timberland owners, forest management organizations, logging and transportation providers, and wood-processing mills - face urgent salvage decisions under uncertain and spatially heterogeneous timber supply conditions. This study presents a scenario-based stochastic optimization framework for post-hurricane salvage logging that explicitly accounts for uncertainty in timber volume loss and quality degradation. Using Hurricane Michael (2018) as a case study in the southeastern United States, forest inventory data, remotely sensed damage indices, species-specific yield curves, and market information are integrated into a spatial optimization model that jointly determines salvage harvest locations and wood transportation decisions. Uncertainty in salvageable supply is represented through multiple simulated damage scenarios with varying levels of sawtimber downgrade to pulpwood. The framework compares risk-neutral and risk-averse strategies, including a mean - risk formulation based on Conditional Value-at-Risk (CVaR), and evaluates sensitivity to key economic and logistical parameters. Results indicate that risk-averse and performance-stable salvage strategies reduce transportation costs and mitigate losses in timber value relative to deterministic approaches, particularly under high uncertainty in damage severity and downgrade rates. Overall, the findings demonstrate that representing post-hurricane timber supply uncertainty through scenario-based simulations and evaluating alternative downgrade and risk preferences materially influence salvage location choices, transportation costs, and economic outcomes. The proposed framework provides a decision-support tool for assessing trade-offs between expected returns and downside risk across a wide range of plausible post-disturbance conditions.
Logging businesses are an essential link within the forest supply chain. A mail survey was conducted to evaluate perspectives on the economic sustainability of the logging industry from across the segments of the forest supply chain in Virginia: logging business owners, consulting foresters (landowner representatives), and mill owners or procurement representatives. More than a third of logging businesses (38.8%) rated the outlook for their business as not economically sustainable. An even higher percentage of consulting foresters and mills also viewed logging operations as not economically sustainable (68.9% and 56.3%, respectively). Sixty percent of logging business respondents were descendants of logging families, and 56% of logging business owners would not encourage their child to enter the logging or forest industry. A smaller percentage of mill representatives (44.9%) and consulting foresters (32.6%) would not encourage their own child to pursue a career in logging or the forest industry. Results indicate economic sustainability challenges could impact recruitment of the next generation of logging business owners. However, the importance of a healthy logging industry was communicated by mill and consulting foresters who expect that in the next five years, their businesses will experience a negative effect due to a lack of logging capacity available to harvest and deliver wood.
