Description and evaluation of Prehas software for pre-harvest assessment of timber assortments

Malinen, Jukka; Kilpeläinen, Harri; Ylisirniö, Kalle

doi:10.1080/14942119.2014.902686

Cited by 6 publications

(21 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Siipilehto et al (2016) noticed some especially large tree height estimates due to the last cut diameter being beyond 10 cm in the used harvester data. Optionally, total tree height (including stump) imputation for STM data could be done by fitting Laasasenaho's (1982) taper curve (Räsänen et al 1998;Malinen et al 2014). Note that possible errors in stump height may also have an effect on tree height and on saw log recovery (see Korhonen et al 2008).…”

Section: Discussionmentioning

confidence: 99%

Model for diameter distribution from assortments volumes: theoretical formulation and a case application with a sample of timber trade data for clear-cut sections

Siipilehto¹,

Rajala²

2019

Silva Fenn.

View full text Add to dashboard Cite

This study examined a theoretical model for stand structures from the volumes of pulpwood and saw logs of clear-cut stands. The average stem size was used to estimate the number of cut trees. The distribution was solved using nonlinear derivative-free optimization. The truncated 2-parameter Weibull distribution was used to describe the stand structure of the commercial stems. This method was first tested with harvester data collected from seven clear-cut stands in southern Finland. Validation included reliability in the stand characteristics and goodness-of-fit of the species-specific distributions. The distributions provided unbiased estimates for the saw log volume, while the bias in the estimated pulpwood volume was 2%. The standard stand characteristics from the Weibull distributions corresponded notably well with the harvester data. A Kolmogorov-Smirnov (KS) test rejected two distributions out of 21 cases, when the accurate input variables were available for the theoretical model. The results of the study suggest that the presented method is a relevant option for predicting the stand structure. In practice, the reliability of the presented method was dependent on the quality of the information available from the stand prior to cutting. With a timber trade data set, the solution for the distribution for a clear-cut section was found. The goodness-of-fit was dependent on the accuracy of the visually assessed timber trade variables. Especially the average stem size proved difficult to assess due to high number of understorey pulpwood stems. Due to overestimated average stem sizes, the solved number of harvested trees was underestimated. Less than 50% of the distributions predicted for clear-cut sections passed the KS test.

show abstract

Section: Discussionmentioning

confidence: 99%

Model for diameter distribution from assortments volumes: theoretical formulation and a case application with a sample of timber trade data for clear-cut sections

Siipilehto¹,

Rajala²

2019

Silva Fenn.

View full text Add to dashboard Cite

show abstract

“…Target stand variables for the ARVO predictions were calculated according to measured plotwise and standwise data. The stem database used in stem group predictions was based on ARVO's pre-installed stem database including stm-data from 423 stands throughout Finland (Malinen et al 2014). The k-MSN method applied in the ARVO tree stock predictions is similar to the MSN method, apart from that k nearest neighbours is selected instead on the nearest neighbour, and the estimates are calculated according to these k nearest neighbours.…”

Section: Evaluation Of the Predicted External Qualitymentioning

confidence: 99%

“…In the ARVO software, the number of the nearest neighbours (k) is set to 10. For more detailed description of the k-MSN method in the ARVO software, see Malinen et al (2014).…”

Section: Evaluation Of the Predicted External Qualitymentioning

confidence: 99%

“…Moreover, TLS (Kankare et al 2016) as well as Mobile Laser Scanning (MLS) or Unmanned Aerial Vehicles (UAV) (Jaakkola et al 2017) may provide adequate reference data in the future. Malinen et al (2014) presented a software, Prehas, capable for assessing the amount and value of harvestable timber, including the predictions of timber assortment recoveries, length-diameter distributions of logs and value recovery of timber. It is based on the utilization of previously collected stem data (stm-data) from CTL harvesters, and application of non-parametric k Most Similar Neighbour (k-MSN) method (Moeur and Stage 1995) and computer bucking simulator.…”

Section: Introductionmentioning

confidence: 99%

“…It is based on the utilization of previously collected stem data (stm-data) from CTL harvesters, and application of non-parametric k Most Similar Neighbour (k-MSN) method (Moeur and Stage 1995) and computer bucking simulator. ARVO, the Finnish version of the Prehas software, includes also an external stem quality database containing more than 13 000 measured sample trees, which is used as a reference data for non-parametric external quality predictions that are utilized in the bucking simulator (Malinen et al 2014). The applicability of stem databases, which include stem dimensions, together with non-parametric methods has been proven in previous studies (Malinen et al 2001;Malinen 2003;Holmgren et al 2012;Barth and Holmgren 2013), but the usability of extensive external quality database has not been evaluated.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Validating the predicted saw log and pulpwood proportions and gross value of Scots pine and Norway spruce harvest at stand level by Most Similar Neighbour analyses and a stem quality database

Malinen¹,

Kilpeläinen²,

Verkasalo³

2018

Silva Fenn.

Self Cite

View full text Add to dashboard Cite

Detailed pre-harvest information about the volumes and properties of growing stocks is needed for increased precision in wood procurement planning for just-in-time wood deliveries by cut-to-length (CTL) harvesters. In the study, the non-parametric Most Similar Neighbour (MSN) methodology was evaluated for predicting external quality of Scots pine and Norway spruce, expressed as stem sections fulfilling the saw log dimension and quality requirements of Finnish forest industry, as they affect the recovery of timber assortments and the value of a pre-harvest stand. Effects of external tree quality were evaluated using saw log recovery and saw log reduction caused by stem defects, as well as total timber value (â¬) and average unit value (â¬ m) in a stand. Root mean square error (RMSE) of saw log recovery and reduction were 9.12 percentile points (pp) for Scots pine and 6.38 pp for Norway spruce stands. In the unit value considerations, the predictions compared with measurements resulted in the RMSE of 3.50 â¬ m and the bias of 0.58 â¬ m in Scots pine stands and 2.60 â¬ m, and 0.35 â¬ m in Norway spruce stands, respectively. The presented MSN based approach together with the utilization of the external stem quality database included in the ARVO software could provide dimension and external quality predictions usable for pre-harvest assessment of timber stock at a stand level. This prediction methodology is usable especially in analyses where timber assortment recoveries, values and unit prices are compared when different bucking objectives are used.â3â3â3â3â3

show abstract

Valuation of growing stock using multisource GIS data, a stem quality database, and bucking simulation

et al. 2018

Self Cite

View full text Add to dashboard Cite

Customer-oriented production as a sawmill strategy requires up-to-date information on the available raw material resources. Bucking is a process in which the tree stem is divided into products based on the roundwood user’s needs regarding products and their quality and dimensions. Optimization methods are employed in bucking to recover the highest value of the stem for a given product price matrix and requested length–diameter distribution. A method is presented here for assessing the value of harvestable timber stands based on their product yield. Airborne laser scanning, multispectral imagery, and field plots were used to produce timber statistics for a grid covering the target area. The statistics for the plots were generated from this grid. The value of the estimated tree list was assessed using a bucking-to-value simulator together with a stem quality database. Different product yield simulations in terms of volumes, timber assortment recoveries, wood paying capabilities (WPC) and value estimations based on the presented method, and extensive field measurements were compared. As a conclusion, this method can estimate WPC for pulpwood and sawlogs with root mean squared errors of 32.7% and 38.5%, respectively, relative to extensive field measurements.

show abstract

Description and evaluation of Prehas software for pre-harvest assessment of timber assortments

Cited by 6 publications

References 12 publications

Model for diameter distribution from assortments volumes: theoretical formulation and a case application with a sample of timber trade data for clear-cut sections

Model for diameter distribution from assortments volumes: theoretical formulation and a case application with a sample of timber trade data for clear-cut sections

Validating the predicted saw log and pulpwood proportions and gross value of Scots pine and Norway spruce harvest at stand level by Most Similar Neighbour analyses and a stem quality database

Valuation of growing stock using multisource GIS data, a stem quality database, and bucking simulation

Contact Info

Product

Resources

About