Models for predicting product recovery using selected tree characteristics of black spruce

2005 ◽  
Vol 35 (4) ◽  
pp. 930-937 ◽  
Author(s):  
Chuangmin Liu ◽  
S Y Zhang
Keyword(s):  
Regression Models ◽  
Black Spruce ◽  
Tree Height ◽  
Product Recovery ◽  
Second Order ◽  
Statistical Regression ◽  
Ann Model ◽  
Back Propagation Algorithm ◽  
Order Polynomial ◽  
Tree Characteristics

The artificial neural network (ANN) model and five traditional statistical regression models were used to predict four parameters of simulated product recovery (lumber volume, lumber value, chip volume, and total product value) from the stud mill simulation based on three basic tree characteristics of black spruce (i.e., diameter at breast height (DBH), tree height, and tree taper). The ANN model (i.e., the three-layer perceptron with error back-propagation algorithm) performed as well as or better than the five statistical regression models in terms of statistical criteria such as R2, root mean square error, and mean absolute error of predictions. The second-order polynomial with both DBH and tree height predicted the four product recoveries as accurately as the ANN model. This study showed that the ANN model, the second-order polynomial function, and the power function were suitable for the prediction of product recovery using the selected tree characteristics. The models developed in this study allow the estimation of the product recovery of individual trees and of a forest stand before it is harvested. It is evident that these models would be valuable tools for forest resource managers.

scholarly journals Equations for predicting tree height, total volume, and product recovery for black spruce (Picea mariana) plantations in northeastern Quebec

2005 ◽  
Vol 81 (6) ◽  
pp. 808-814 ◽  
Author(s):  
Chuangmin Liu ◽  
S Y Zhang
Keyword(s):  
Black Spruce ◽  
Management Strategies ◽  
Tree Height ◽  
Product Recovery ◽  
Tree Level ◽  
Stem Volume ◽  
Model Parameters ◽  
Total Product ◽  
Tree Product ◽  
Tree Characteristics

Several regression models with different independent variables were studied for their ability to predict total tree height, total stem volume, and product recoveries (lumber volume, chip volume, lumber value, and total product value) from a sawing simulator. A sample of 172 trees from black spruce plantations was used to fit model parameters and another independent sample of 139 trees was used for model evaluation. The sample encompassed large variations in tree characteristics and tree product recovery. All the fitted models were suitable for predicting their corresponding response variables. Model validation through actual product recovery data from a real stud mill further indicated that the general tree-level models for the product recovery were able to accurately predict product recovery, especially from small- and medium-sized trees, using measured tree characteristics. These models provide a valuable tool for forest managers in determining appropriate management strategies (e.g., stand volume and optimizing stand value). Key words: black spruce, regression analysis, tree characteristics, product recovery, sawing simulation

Modelling the relationship of tree-level product value with tree characteristics in black spruce

2006 ◽  
Vol 82 (5) ◽  
pp. 690-699 ◽  
Author(s):  
S Y Zhang ◽  
Y C Lei ◽  
Z H Jiang
Keyword(s):  
Black Spruce ◽  
Tree Height ◽  
Tree Level ◽  
Elasticity Analysis ◽  
Scale Elasticity ◽  
Stem Taper ◽  
Product Value ◽  
Relationship Of ◽  
The Relationship ◽  
Tree Characteristics

The establishment of the relationship between tree-level product value and tree characteristics will allow for predicting the potential value of individual trees and a stand directly using tree characteristics. Using statistical and elasticity analysis methods this study examined the relationship of tree-level product value with selected tree characteristics in black spruce (Picea mariana). The study was based a sample of 139 trees from 48-year-old black spruce plantations grown in Ontario, Canada. The sample trees showed large variation in tree characteristics and tree-level product value. Models were developed and compared on the basis of statistics of the estimated and predicted criteria. Results show that the model, including only tree DBH, tree height and stem taper, is the best in describing the relationship of the tree-level product value with tree characteristics. Furthermore, relationships including input-output and interaction factors in the model were analyzed by calculating the elasticity of production and scale and the cross partial derivative of output with respect to the inputs. The analyses indicate that tree DBH has the largest and positive influence on tree-level product value, followed by tree height; however, stem taper has a negative effect on tree-level product value. When tree DBH, tree height and stem taper each increase by 1%, the quantities of output elasticity show 2.53%, 0.64% and -0.37% changes in the product value, respectively; while the scale elasticity shows a 2.81% increase in tree-level product value with a simultaneous 1% change in tree DBH, tree height and stem taper. Results indicate that the model is suitable for predicting tree-level product value using those tree characteristics from forest inventory and also reflects biological behaviour.Key words: black spruce, regression models, elasticity analysis, product value, tree characteristics

Modelling knot morphology as a function of external tree and branch attributes

2013 ◽  
Vol 43 (3) ◽  
pp. 266-277 ◽  
Author(s):  
E. Duchateau ◽  
F. Longuetaud ◽  
F. Mothe ◽  
C. Ung ◽  
D. Auty ◽  
...  
Keyword(s):  
Tree Growth ◽  
Pinus Banksiana ◽  
Black Spruce ◽  
Three Dimensional ◽  
Tree Height ◽  
Jack Pine ◽  
Weibull Function ◽  
Data Set ◽  
External Branch ◽  
Tree Characteristics

Existing models for describing knot morphology are typically based on polynomial functions with parameters that are often not biologically interpretable. Hence, they are difficult to integrate into tree growth simulators due to the limited possibilities for linking knot shape to external branch and tree characteristics. X-ray computed tomography (CT) images taken along the stems of 16 jack pine (Pinus banksiana Lamb.) trees and 32 black spruce (Picea mariana (Mill.) B.S.P.) trees were used to extract the three-dimensional shape of 3450 and 11 276 knots from each species, respectively. Using a nonlinear approach, we firstly fitted a model of knot geometry adapted from a Weibull function. Separate equations were used to describe both the curvature and the diameter of the knot along its pith. Combining these two equations gave an accurate representation of knot shape using only five parameters. Secondly, to facilitate the integration of the resulting model into a tree growth simulator, we extracted the parameters obtained for each knot and modelled them as functions of external branch and tree characteristics (e.g., branch diameter, insertion angle, position in the stem, tree height, and stem diameter). When fitted to a separate data set, the model residuals of the black spruce knot curvature equation were less than 2.9 mm in any part of the knot profile for 75% of the observations. The corresponding value from the diameter equation was 2.8 mm. In jack pine, these statistics increased to 5.4 mm and 3.2 mm, respectively. Overall, the ability to predict knot attributes from external tree- and branch-level variables has the potential to improve the simulation of internal stem properties.

scholarly journals Allometric Equations for Estimating Biomass and Carbon Stocks in Afforested Open Woodlands with Black Spruce and Jack Pine, in the Eastern Canadian Boreal Forest

Forests ◽  
2021 ◽  
Vol 12 (1) ◽  
pp. 59
Author(s):  
Olivier Fradette ◽  
Charles Marty ◽  
Pascal Tremblay ◽  
Daniel Lord ◽  
Jean-François Boucher
Keyword(s):  
Large Scale ◽  
Ad Hoc ◽  
Black Spruce ◽  
Tree Height ◽  
Jack Pine ◽  
Carbon Stocks ◽  
Allometric Equations ◽  
C Stocks ◽  
The Difference ◽  
Canadian Boreal Forest

Allometric equations use easily measurable biometric variables to determine the aboveground and belowground biomasses of trees. Equations produced for estimating the biomass within Canadian forests at a large scale have not yet been validated for eastern Canadian boreal open woodlands (OWs), where trees experience particular environmental conditions. In this study, we harvested 167 trees from seven boreal OWs in Quebec, Canada for biomass and allometric measurements. These data show that Canadian national equations accurately predict the whole aboveground biomass for both black spruce and jack pine trees, but underestimated branches biomass, possibly owing to a particular tree morphology in OWs relative to closed-canopy stands. We therefore developed ad hoc allometric equations based on three power models including diameter at breast height (DBH) alone or in combination with tree height (H) as allometric variables. Our results show that although the inclusion of H in the model yields better fits for most tree compartments in both species, the difference is minor and does not markedly affect biomass C stocks at the stand level. Using these newly developed equations, we found that carbon stocks in afforested OWs varied markedly among sites owing to differences in tree growth and species. Nine years after afforestation, jack pine plantations had accumulated about five times more carbon than black spruce plantations (0.14 vs. 0.80 t C·ha−1), highlighting the much larger potential of jack pine for OW afforestation projects in this environment.

scholarly journals Using Artificial Neural Networks in Predicting the Level of Stress among Military Conscripts

Mathematics ◽  
2021 ◽  
Vol 9 (6) ◽  
pp. 626
Author(s):  
Svajone Bekesiene ◽  
Rasa Smaliukiene ◽  
Ramute Vaicaitiene
Keyword(s):  
Neural Network ◽  
Transfer Functions ◽  
Back Propagation ◽  
Armed Forces ◽  
Activation Function ◽  
Optimal Number ◽  
Ann Model ◽  
Back Propagation Algorithm ◽  
Classification Rate ◽  
Artificial Neural

The present study aims to elucidate the main variables that increase the level of stress at the beginning of military conscription service using an artificial neural network (ANN)-based prediction model. Random sample data were obtained from one battalion of the Lithuanian Armed Forces, and a survey was conducted to generate data for the training and testing of the ANN models. Using nonlinearity in stress research, numerous ANN structures were constructed and verified to limit the optimal number of neurons, hidden layers, and transfer functions. The highest accuracy was obtained by the multilayer perceptron neural network (MLPNN) with a 6-2-2 partition. A standardized rescaling method was used for covariates. For the activation function, the hyperbolic tangent was used with 20 units in one hidden layer as well as the back-propagation algorithm. The best ANN model was determined as the model that showed the smallest cross-entropy error, the correct classification rate, and the area under the ROC curve. These findings show, with high precision, that cohesion in a team and adaptation to military routines are two critical elements that have the greatest impact on the stress level of conscripts.

AN ARTIFICIAL INTELLIGENT APPROACH TO TRAFFIC ACCIDENT ESTIMATION: MODEL DEVELOPMENT AND APPLICATION

Transport ◽  
2009 ◽  
Vol 24 (2) ◽  
pp. 135-142 ◽  
Author(s):  
Ali Payıdar Akgüngör ◽  
Erdem Doğan
Keyword(s):  
Genetic Algorithm ◽  
Model Development ◽  
Back Propagation ◽  
Linear Functions ◽  
Model Parameters ◽  
Ann Model ◽  
Back Propagation Algorithm ◽  
Estimation Model ◽  
Average Growth ◽  
Feed Forward Back Propagation

This study proposes an Artificial Neural Network (ANN) model and a Genetic Algorithm (GA) model to estimate the number of accidents (A), fatalities (F) and injuries (I) in Ankara, Turkey, utilizing the data obtained between 1986 and 2005. For model development, the number of vehicles (N), fatalities, injuries, accidents and population (P) were selected as model parameters. In the ANN model, the sigmoid and linear functions were used as activation functions with the feed forward‐back propagation algorithm. In the GA approach, two forms of genetic algorithm models including a linear and an exponential form of mathematical expressions were developed. The results of the GA model showed that the exponential model form was suitable to estimate the number of accidents and fatalities while the linear form was the most appropriate for predicting the number of injuries. The best fit model with the lowest mean absolute errors (MAE) between the observed and estimated values is selected for future estimations. The comparison of the model results indicated that the performance of the ANN model was better than that of the GA model. To investigate the performance of the ANN model for future estimations, a fifteen year period from 2006 to 2020 with two possible scenarios was employed. In the first scenario, the annual average growth rates of population and the number of vehicles are assumed to be 2.0 % and 7.5%, respectively. In the second scenario, the average number of vehicles per capita is assumed to reach 0.60, which represents approximately two and a half‐fold increase in fifteen years. The results obtained from both scenarios reveal the suitability of the current methods for road safety applications.

scholarly journals Second-order statistical regression and conditioning of replicate transient kinetic data

2008 ◽  
Vol 63 (7) ◽  
pp. 1850-1865 ◽  
Author(s):  
Raf Roelant ◽  
Denis Constales ◽  
Roger Van Keer ◽  
Guy B. Marin
Keyword(s):  
Kinetic Data ◽  
Second Order ◽  
Statistical Regression
Sign in / Sign up

Export Citation Format

Share Document