Printer Friendly

Evaluating the system logistics of a centralized biomass recovery operation in Northern California.

Abstract

In this study, we evaluated the productivity and cost of each component in a unique centralized biomass recovery operation to determine cost-effective system logistics. The system was divided into three segments: collection, comminution, and transportation. Cost analysis determined a stump-to-truck and transportation cost of $30.39 and $13.91 per bone dry ton (BDT), respectively. These costs did not include support equipment, overhead, profit, or risk allowance. Transportation cost was evaluated for a total one-way haul distance of 15 miles. To control overall cost it is imperative to maintain maximum productivity of processing, our most expensive component. Therefore, upstream and downstream practices were examined to determine how they influenced the system. Analysis of the loader operation showed a 33 percent increase in cycle time when handling hardwood whole trees piled at a landing, compared with conifer slash piled within the unit. Regression analysis of the modified dump truck used in the prehaul confirmed that distance had a significant impact on overall centralized biomass grinding operations. Sensitivity analysis showed that a 20 percent reduction in productivity owing to increased travel time resulted in a 25 percent increase in grinding cost. A decoupled transport system used all-wheel drive tractors to haul comminuted biomass from the grinder to a trailer landing, where regular highway tractors completed the trip to the power plant. This system reduced delays in loading and improved access to the grinding location. Through an understanding of the complete system, a manager could identify cost saving elements, adjust upstream productivity to meet demand, and reduce the overall cost of a centralized biomass recovery operation.

**********

The use of forest residues has great potential for expanding biomass electric power generation on the north coast of California. Currently, processed submerchantable trees, limbs and tops from processing trees into sawlogs, and mill waste supply a third, or 60 megawatts (MW), of Humboldt County's electric demand (Schatz Energy Research Center 2013). The availability of woody biomass within the county has been estimated to provide enough to support up to 220 MW (Williams et al. 2007).

In addition to increasing the county's power generation, the recovery of woody biomass in the form of logging residues benefits commercial timberlands by significantly reducing site preparation costs. The direct cost to remove piled residues is about the same as burning it without the added risk of an escaped fire and air quality issues (M. Alcorn, Green Diamond Resource Company, personal communication, 2012). In addition, the need to control fire stimulated weed species such as blueblossom (Ceanothus thyrsiflorus) and manzanita (Arctostaphylos manzanita) with herbicides is reduced. The total cost savings for reforestation activities can range from $350 to $800 per acre or more when consequent reductions in carbon emissions, fire risks, and herbicide application are considered in biomass recovery (M. Alcorn, personal communication, 2012).

Benefits from biomass recovery can also be realized in the mechanical removal of slash piles created from fuel management practices. This method has been used in National Forests in northern California as an alternative to open pile burning, avoiding negative effects such as smoke production, residual tree mortality, and the risk of fire escape (Han et al. 2010). It also reduces the time required to proceed to the next step of understory burning, which often follows pile burning.

Despite these benefits, forest residues remain underused because of economic and operational barriers related to the cost of collecting, processing, and transporting a product with low market value (Han et al. 2002). Pan et al. (2008) reported a total production cost to feller bunch, skid, and grind small-diameter trees (<5.0 in.) at $29.16 per bone dry ton (BDT). Productivity of primary transport of whole trees to the landing over an average distance of 884 feet was 17.41 BDT per productive machine hour (PMH) at a cost of $6.08/BDT. To improve productivity and reduce costs for the whole system, they suggested reducing operational delays by matching the productivities of each machine in the operation or "achieve system balance." In addition, costs could also be reduced if residues piled at the landing from other logging operations were used.

A different study evaluated the prehaul of loose residues densified into bundles from a landing to a centralized grinding area using hook-lift trucks. The total production cost for this operation was $60.98/BDT. The productivity of loading and hauling bundles 4.5 miles to a grinder was 10.07 BDT/PMH at a cost of $16.62/BDT. Unfortunately, the bundler and loader were not decoupled, which resulted in a poor system balance. An improvement in the pairing of machine capacities and operational productivity could have resulted in a reduction of cost (Harrill et al. 2009).

In a similar study, hook-lift trucks were used to haul loose logging slash to a centralized location for processing. The total production cost to load, haul, and grind was $32.98/ BDT. The difference in productivity of loading (20.10 BDT/ PMH) and hauling (9.93 BDT/PMH) the loose slash led to bottlenecks in the operation. To reduce costs, the authors suggested carefully planning a strategic logistical arrangement of machines with multiple hook-lift trucks at a minimal travel distance (Harrill and Han 2010).

A hook-lift truck used to prehaul hand-piled slash from shaded fuel break treatments inaccessible by typical highway chip trucks was also studied. Collection and prehauling to a central grinding site costs were $31.25/ BDT. It should be noted that prehaul costs were sensitive to haul distance because of slow travel speeds on winding, mountainous roads and low slash weight per turn (Han et al. 2010).

In a more recent study (Anderson et al. 2012), the transportation of unprocessed slash to a concentration yard using fifth wheel end-dump trucks was compared with an inwoods grinding system using high-sided dump trucks to haul ground material to a concentration yard. System costs were comparable at $23.62 and $24.52 per BDT, respectively. In both cases, the system was dependent on appropriately balancing productivity rates of individual machines.

In all of these studies, an understanding of individual machine productivity and cost was essential in determining an efficient system balance. To improve the knowledge of cost reduction methods and system balance techniques, our study investigated the production cost of collecting, processing, and transporting forest residues from mixed conifer even-age harvests on the north coast of California. The emphasis of this study was to identify key variables that influenced the productivity of individual components of the centralized biomass recovery system. These findings could then be used to guide recommendations on ways to establish and manage system balance in an effort to control cost.

Methods

Study site and biomass recovery system description

The study was conducted on private industrial timberland in northern California that typically uses even-aged management. Forest residues in the form of limbs, tops, small-diameter trees, and submerchantable trees of mixed conifer and hardwood were piled along the roadside, within units, or scattered throughout the harvested units. Six harvested units ranging from 7 to 33 acres in size were selected for the study. Biomass recovery operations in all units followed the same system logistics. The vegetation was second growth mixed conifer consisting of redwood (Sequoia sempervirens), Douglas-fir (Pseudotsuga menziesii), and tanoak (Notholithocarpus densiflorus). The units were previously harvested using either a shovel logging or cable yarding system, depending on ground slope. The amount of biomass and location of piles was dependent on the harvest method used. Biomass from processing trees was typically piled in the unit for shovel logging systems. Biomass from cable yarded units was piled at the landing. We characterized the biomass material into three groups: slash, whole trees, or mixed. Slash was typically limbs and tops generated from processing logs. Whole trees were either nonmerchantable conifer or hardwood, such as tanoak. The amount of biomass removed from a unit ranged from 12 to 55 BDT/acre and was dependent on harvest unit size, harvest method, and access.

The study observed the collection, comminution (i.e., grinding), and transportation of forest residues (Fig. 1). The collection segment of the operation started in the harvested unit with a loader (Linkbelt 3400) using a rotating 10-tine grapple. The loader collected biomass from piles at roadside landings, piles within the unit, or scattered residues. Its primary function was to load an all-wheel drive (AWD) articulated dump truck (Volvo A35C or Caterpillar D300D) modified with additional side walls and a rear gate extension to increase the normal carrying volume to 50 cubic yards or more (Fig. 2). In addition, the dump trucks were outfitted with skidder tires for increased traction when driving off spur roads into the harvested unit. This allowed the truck to haul material over native surface spurs and single-lane roads with an average grade of [+ or -] 4 percent to a centralized grinding site.

The comminution segment was the core of the centralized biomass recovery operation. It incorporated a majority of the machines, beginning with the dump truck delivering material to a loader. The loader (Linkbelt 3400) used a rotating seven-tine grapple to swing dumped material 90[degrees] to 180[degrees] onto the grinder's (Peterson Pacific 5710C) infeed conveyer. After processing, the ground material was fed via conveyor into a positioned chip trailer. This segment of the operation required adequate space (approximately 10,000 [ft.sup.2]) to accommodate all machines, including passing log trucks from other operations (S. Morris, Steve Morris Logging, personal communication, 2013).

The transportation segment was decoupled using tractors to haul chip trailers from the centralized grinding area to a trailer transfer site, where additional tractors would haul the material the rest of the distance to the power plant. AWD tractors, modified from cement trucks (Kenworth, Peterbilt, and Oshkosh), were used to haul loaded chip trailers an average of 2 miles, each way, to the transfer site. Multiple trucks were used to ensure minimal delay in the grinder's hot operation. The trailer transfer sites were typically located roadside on well-traveled, two-lane, rocked roads. The final haul to the power plant was done using regular highway tractors. The total observed one-way travel from stump to power plant was 15 miles.

Data collection and analysis

Hourly cost for each machine was calculated using standard methods (Miyata 1980, Brinker et al. 2002). Machine rate use factored in the cost to purchase and modify used machines (Table 1). Detailed time study data were collected to estimate machine productivity and delays using standard harvesting work study techniques (Olsen et al. 1998). Time duration of cycle elements and delays for each machine were recorded using centiminute (min/100) stop watches. Independent variables hypothesized to have an influence on machine productivity were also recorded for each cycle. Data were screened for outliers and predictive equations were developed using ordinary least squares regression. Normality was tested using the Shapiro-Wilk test. Homoscedasticity was confirmed by visually examining residual plots. Multicollinearity within the data was tested using both condition numbers and variance inflation factors. Parameters with P < 0.05 were considered significant. Calculations were performed using the R 2.15.1 statistical software program (R Core Team 2012).

Moisture content of samples collected from filled chip trailers was used to calculate weight in BDT and determine production rates (BDT/PMH). Moisture content was calculated using a modified version of the American Society of Agricultural Engineers (ASAE) Standard $358.2 protocols (ASAE 2003). Samples were collected by filling a 5-gallon bucket from three locations along the top of a chip trailer. The sample was mixed, and a 1-gallon subsample was taken for analysis. Samples were dried at 103[degrees]C for 48 hours, as opposed to 24 hours per protocol. The samples were reweighed to determine wet basis moisture content.

Calculations for transportation of slash and ground material requiring distance, rate of speed, and road grade were collected using a hand-held Garmin Csx60 GPS unit and analyzed using ArcMap 10.0 (Environmental Systems Research Institute 2011).

Loading modified dump trucks

The loader's activities included swinging empty, grappling, and swinging loaded. The swinging empty cycle began when the arm swung away from the dump truck bed and stopped when it made contact with material. Grappling was timed from first contact to when the arm began its swing to the dump truck. Swinging loaded began when the arm swung toward the dump truck and stopped over the dump bed. Pile type, location of pile, species, and arm swing degree were estimated visually as predictors for regression analysis. Time associated with movement to a new pile and collecting material was also recorded to evaluate its contribution to productivity. Compacting or sorting material was not recorded because this activity rarely occurred. To calculate productivity, the average BDT per dump truck was divided by the average time to load a dump truck. An average BDT per dump truck was the average BDT per chip trailer obtained from scale tickets divided by the average number of dump truck loads needed to fill a chip trailer.

Dump trucks prehauling forest residues to a centralized grinding site

The round-trip dump truck cycle elements include traveling empty, positioning empty, loading, traveling loaded, positioning loaded, and unloading. Traveling empty began when the dump bed returned to the lowered position as it pulled away from the centralized grinding area and ended when it stopped to position itself. Positioning empty started when it began backing into position and concluded when it stopped next to the loader. This is when loading began, and it ended when the dump truck pulled away. Traveling loaded ended when the dump truck stopped to position for unloading. Positioning loaded started when the truck began to back into position and ended when it stopped next to the loader. Unloading began when the truck stopped backing up and ended when the dump bed returned to the lowered position. Slope and distance were recorded when traveling and positioning. Material type (i.e., whole tree or slash) loaded into the truck was also recorded.

Loader and grinder in the centralized grinding site

Productivity of the loader and grinder, positioned at the centralized grinding area, was calculated by dividing the average BDT per chip trailer obtained from scale tickets by the average time to fill one chip trailer. This production rate was assumed to be equal for both machines because any delay in the loader would directly cause a delay with the grinder.

AWD tractor and chip trailer shuttling containers

The AWD tractor's cycle elements included traveling loaded, positioning loaded, unhooking, traveling to empty trailer, positioning to hook up trailer, hooking up trailer, traveling empty, positioning empty, and loading. The traveling loaded cycle began when the truck pulled away from the grinder and ended when it came to a stop to position for unloading. This was the beginning of positioning loaded, which ended when the truck stopped moving. Unhooking started when the truck stopped positioning and lasted until it pulled away without a trailer. Traveling to the new trailer ended when it stopped in front of the empty trailer. Positioning to hook up trailer finished when it stopped underneath the new trailer. This started the hooking cycle and lasted until the truck pulled away with the empty trailer. Traveling empty ended when it stopped for positioning empty. Positioning empty finished when it stopped in position to pull in front of the grinder. Loading began when the trailer pulled underneath the grinder's conveyor. Other variables recorded were distance of each travel segment, average road slope, and weights from scale tickets.

Highway tractor and chip trailer hauling ground material to a power plant

The cycle elements recorded were similar to the AWD chip truck except when the truck was unloading at the power plant. Unloading started when the truck stopped on the truck tipper and ended when the tipper returned to level position. The three road classes traveled (i.e., double-lane rock, two-lane paved, or highway) were also recorded.

Total system cost

Individual productivity rates were calculated for each machine to determine its contribution to the system. The system cost of this hot operation was based on the grinder and loader as the limiting production rate. When production of other machines fell below this limit, more machines were added. Therefore, the sum of costs of all machines ($/BDT) was divided by the limiting production rate (BDT/PMH) to get the system cost.

Results and Discussion

Moisture content analysis

Moisture content samples were collected from 22 trucks. The average green weight of a chip van load was 22.5 tons (SD = 1.8 tons) with an average wet basis moisture content of 25 percent (SD = 7%). Variation in moisture content could be a result of varying material type, species, or environmental conditions during the operation.

Loading activity analysis

Three primary activities were observed with the loader: loading the dump trucks, compiling scattered material, and repositioning to a new pile. The average time spent loading a dump truck was 4.74 minutes using an average of 10 grapple loads. The average time spent collecting material was 3.77 minutes using an average of 9.59 grapples. The total delay-free time recorded consisted of collecting (48%), grappling material (21%), swinging loaded (18%), and swinging empty (13%). Occasionally the loader would reposition. On average this time took less than 0.30 minute. The productivity of the loader averaged 46.57 BDT/PMH with a production cost of $5.25/BDT. This productivity is comparable to that found by Harrill et al. (2009) in the Sierra Nevada Mountains when loading densified bundles of slash into hook-lift trucks.

Regression analysis was used to see whether loaded swing arc degree, empty swing arc degree, pile type (whole tree or slash), pile location (in the unit, roadside, or not piled but scattered throughout the unit), and pile species (hardwood, conifer, or a mix) had a significant effect on cycle time (Table 2). Despite efforts to reduce outliers and influential observations in the data, the model generated had a low [r.sup.2] value. The model's inability to explain variation does limit its ability to accurately predict, but can provide an inference on trends in the data. Whole tree piles were found to increase cycle time compared with slash. This may be because the size and length of whole trees make them more difficult to handle. Pile location also had an influence on cycle time. Material collected in the unit influenced cycle time the least compared with roadside material. This may be a result of the type of material handled at each location. Roadside material was typically whole trees, which tend to be more difficult to handle and load compared with loose slash pile in units. Scattered material increased cycle time the greatest. Cases with mixed material were not found to be significant (P = 0.81). In general the harvest system used determined the locations of piled material. In ground-based units, whole tree hardwoods and mixed conifer slash was either piled on the roadside, piled within the unit, or scattered throughout. Cable units typically had piles along the roadside or landings. Our findings suggest that how and where residuals are left can have an impact on future biomass recovery operations, because accessibility and organizing material into piles were found to be the greatest influences on cycle time. In addition, harvesting method can be a major factor affecting the amount and location of material.

Dump truck prehauling operations

The dump truck round-trip cycle time averaged 12.69 minutes traveling an average one-way distance of 0.48 mile on an average grade of [+ or -] 4 percent. It traveled at an average speed of 9.30 miles per hour (mph) carrying 5.63 BDT per trip. Productivity averaged 26.16 BDT/PMH, resulting in a production cost of $4.42/BDT. This rate is more than double the production reported from both Harrill et al. (2009) hauling bundles with hook-lift trucks and Harrill and Han (2010) hauling loose slash with the same types of trucks. At this rate, a single dump truck could not meet the grinder's production. For this reason, two to four dump trucks were used depending on travel distance.

On average, a delay-free cycle consisted of loading (34%), traveling loaded (23%), traveling empty (20%), positioning empty (13%), positioning loaded (4%), and unloading (5%). The difference in time when traveling loaded compared with empty was attributed to slower travel speeds to avoid material falling off the back, especially on steeper road grades. Slower speeds were also noticed when empty dump trucks backed down single-lane spur roads to position for loading. Long positioning distance due to lack of turn around locations also increased cycle time.

Regression analysis did not show any significant variables (P < 0.05) influencing cycle time when the model included traveling empty or loaded distances, slope, or material type (Table 3). However, a reduced model regressing delay-free time by total distance traveled was highly significant (P < 0.05; Table 4). A sensitivity analysis using the reduced regression model shows production costs starting at $2.74/ BDT at a one-way distance of a quarter mile with an increase of $0.83/BDT for every additional quarter mile.

Centralized grinding productivity and cost

The rate of production was assumed to be the same for the loader and grinder. The two machines worked as a team, and any delay in the loader's production would have directly affected the grinder. The grinder processed an average of 17.15 BDT in 26.05 minutes for an average production rate of 38.04 BDT/PMH. The grinder had the highest machine cost ($448.02/PMH) compared with any other machine in this operation. Production cost was $11.78/BDT for the grinder and $6.10/BDT for the loader, for a total of $17.87/ BDT. Samples taken from chip trailers and analyzed in the laboratory determined a moisture content range of 14 to 38 percent.

Transportation of ground materials to a power plant

The AWD tractor and chip trailer had an average cycle time of 47.99 minutes on an average one-way distance of 2.13 miles. Productivity for the tractor and trailer was 26.47 BDT/PMH. For this reason, at least two tractors were needed to meet grinder production. Tractor and trailer production costs were $4.44/BDT and $0.15/BDT, respectively, for a total of $4.59/BDT. The greatest percentage of time was spent waiting while being loaded (53%). Travel time was 37 percent of overall time. Traveling and positioning empty (15%) was similar to traveling and positioning loaded (12%). This difference was in the extra time needed to position the truck and empty trailer within the limited space of the centralized grinding area. The remaining percentage was for transferring trailers (10%) and waiting to be loaded (9%). Sensitivity analysis shows that production costs increase S0.44/BDT for every additional quarter-mile one-way distance. More data would be needed to develop a regression model and develop further analysis. It should be noted that production cost could significantly increase with an increase in hauling distance. One significant advantage of using modified AWD tractors was their capability of hauling loaded trailers over adverse forest road conditions. This reduced the need to put a centralized grinding site on a main haul road that could be accessed by normal highway tractors. This reduced dump truck prehaul distances and increased the cost-effective range of the operation.

The standard highway tractors used to pick up loaded chip trailers traveled an average of 13 miles one way with a cycle time of 86.34 minutes. The tractors hauled 11.92 BDT/PMH at a combined tractor and trailer cost of $7.53/ BDT. Traveling empty time (29.15 min) was nearly the same as loaded traveling time (30.65 min). The truck's route to the power plant was over two-lane rock and paved roads only. The average speed on the two-lane rock road was 27 mph for a one-way distance of 9.5 miles. The average speed on two-lane paved roads was 31 mph at a distance of 3.4 miles. Compared with AWD, highway tractors had half the production rate even though they pulled the same weight. This is largely because of the difference in transportation distances and speed of travel.

System balance directly affecting productivity and cost

This centralized biomass recovery operation involved a combination of machines to collect, process, and transport material. The balance of production within the system is important because it has a direct effect on the overall production cost. Unused capacity resulting from an imbalanced system will reduce units of output, increasing fixed costs of underused functions, which in turn drives up per-unit production costs (Rummer 2008).

The highest per-unit production cost observed was the grinding operation at $17.87/BDT for both loader and grinder (Table 5). To minimize total production cost, the operation would first need to maximize processing productivity by reducing any delay created between the loader and grinder. Four percent of the total time observed was characterized as operational delay (Table 6). A majority (71%) of this delay represented waiting on the dump truck to deliver more material. Delay in delivering biomass to the grinder could have come from either the loader or the dump truck. Observations of the loader revealed a very efficient operation with less than 1 percent delay overall. The dump truck, on the other hand, experienced delay 17 percent of the total time observed. A little more than half (51%) of this delay occurred when waiting for another dump truck to finish being loaded. The remaining delay was waiting for the loader to compile biomass (43%) and waiting for other operational and personal delays (6%). These findings suggest that compiling scattered material is leading to inefficiencies downstream, and therefore piling material during harvesting activities could reduce costs in biomass recovery operations.

Balancing the collection segment to match grinding production can also be challenging when considering the spatial diversity of material influencing travel time and productivity. To assist in balancing upstream functions, a sensitivity analysis was conducted to calculate the productivity of different truck combinations at various distances. A single dump truck could travel 0.38 mile one way and meet the 38 BDT/PMH grinder demand. At 1.25 miles, its productivity drops to 19 BDT/PMH or 50 percent of the demand, requiring an additional truck to keep the system in balance and the overall operational cost at a minimum. Analysis on the effects of dump truck productivity on the combined loader and grinder cost revealed that a 20 percent drop in dump truck production would result in a 25 percent increase in comminution cost or $4.45/BDT (Fig. 3).

A decoupled transportation system was successfully used to minimize potential delays between the comminution and transportation segments. The "hot" load operation, where material is directly loaded into a trailer while grinding, required a minimum of two chip trucks to ensure an empty one was always available for loading. During the operation, only 2 percent of the total time observed was in delay. The time an empty truck spent waiting for the previous truck to finish being loaded (4.25 min) was not considered delay time. This buffer ensured that a truck was available to load. Managing downstream productivity required matching the right amount of chip trucks to the distance traveled to the transfer location.

The loaded trailer pick up at the transfer location required that an empty trailer be available every 26 minutes. Eight empty trailers were employed to ensure availability. The loaded trailers were taken to one of three local power plants located 13, 31, or 57 miles away. The destination varied depending on the loaded material because each facility had general feedstock demands. Delays of 12 percent while waiting to be unloaded were observed.

The total stump-to-truck cost for this operation was $30.39/BDT. This is greater than the reported $24.52/BDT cost from a similar study investigating high-sided dump trucks used to prehaul material (Anderson et al. 2012). As in Anderson's study, it should be noted that this study's estimated total cost does not include overhead, profit, and risk allowance, nor supporting equipment needed for the operation. The total system cost including transportation was $44.30/BDT. These cost figures can increase significantly with an increase in transportation distance.

Conclusions

This study evaluated the productivity and cost of collecting, comminuting, and transporting biomass to better understand a centralized biomass recovery operation's logistics. Each component in the system was analyzed to identify key variables that influence overall system balance. Productivity for different machines varied from 12 to 47 BDT/PMH with a cost range of $5.87 to $11.78 per BDT. The total stump-to-truck system cost was $30.39/BDT, with an additional $13.91/BDT for transportation to the power plant.

The collection segment of the operation used a loader and two modified dump trucks to prehaul biomass at a cost of $12.51/BDT. The type of material and location influenced loading cycle times. Whole tree material and material scattered within the unit both increased cycle times. In addition, the harvesting system was found to be a major factor affecting the amount and location of residuals. Our findings suggest that how and where a harvest contractor leaves residues can have an impact on a biomass recovery operation.

Prehauling with dump trucks (26.16 BDT/PMH) over an average one-way distance of a half mile was more productive than other methods found in the literature. Analysis varying the amount of truck and one-way travel distance determined that a second truck would be needed at 1.25 miles to meet the grinder's production rate. This type of analysis can be important when designing a work plan to achieve system balance.

The comminution segment was the hub of the operation. It provided the demand that upstream and downstream production had to meet. The loader and grinder produced 38.04 BDT/PMH at a combined cost of $17.87/BDT. The variability of dump truck productivity (owing to distance and number of trucks used) on comminution costs revealed that a 20 percent drop in dump truck production would result in a 25 percent increase in comminution cost or $4.47/ BDT. This is mainly a result of high operating costs for the grinder and suggests that careful planning of prehaul distances can reduce potential delay and control cost.

The transportation segment was successful in limiting comminution delays while directly loading into the chip trailer by using a decoupled system. This system required two to three AWD tractors, a trailer transfer site, and additional highway tractors to haul the remaining distance to the power plant. The AWD tractors were capable of hauling loaded trailers over adverse forest road conditions at a rate of 26.47 BDT/PMH for $6.38/BDT. The final leg of transportation, which traveled a greater distance, had a production rate of 11.92 BDT/PMH with a cost of $7.53/ BDT. Downstream production efficiency was clearly due to a decoupling of the transportation from comminution segment. The use of AWD tractors reduced the need to put a centralized grinding area on a main haul route, which reduced dump truck travel distances and increased the overall extent of biomass recovery operation. In addition, it decoupled the hot operation from downstream transportation, reducing potential delays. A similar system configuration is recommended if the equipment is available.

This study provided a productivity and cost for each machine in a centralized biomass recovery operation using unique equipment. In addition, sensitivity analysis on travel distances for transport machines provided additional information to better understand total production costs, which could be useful to managers when planning similar biomass recovery operations.

Acknowledgments

Funding for this study was provided by the Biomass Research and Development Initiative (BRDI) of the USDA's National Institute of Food and Agriculture (NIFA), with in-kind support from the Green Diamond Resource Co. and Steve Morris Logging.

Literature Cited

Anderson, N., W. Chung, D. Loeffler, and J. G. Jones. 2012. A productivity and cost comparison of two systems for producing biomass fuel from roadside forest treatment residues. Forest Prod. J. 62(3):222-233.

American Society of Agricultural Engineers (ASAE). 2003. $358.2--Moisture measurement--Forages. In: ASAE Standards. 47th ed. ASAE, St. Joseph, Michigan. 595 pp.

Brinker, R., J. Kinard, B. Rummer, and B. Lanford. 2002. Machine rates for selected forest harvesting machines. USDA Forest Service, Southern Research Station, Asheville, North Carolina. 32 pp.

Environmental Systems Research Institute (ESRI). 2011. ArcGIS Desktop: Release 10. ESRI, Redlands, California.

Han, H.-S., J. Halbrook, F. Pan, and L. Salazar. 2010. Economic evaluation of a roll-off trucking system removing forest biomass resulting from shaded fuelbreak treatments. Biomass Bioenergy 34:1006-1016.

Han, H.-S., H. W. Lee, L. R. Johnson, R. L. Folk, and T. M. Gorman. 2002. Economic feasibility of small wood harvesting and utilization on the Boise National Forest--Cascade, Idaho City, Emmett Ranger Districts. College of Natural Resources, Department of Forest Products, University of Idaho, Moscow. 61 pp.

Harrill, H. and H.-S. Han. 2010. Application of hook-lift trucks in centralized slash grinding operations. Biofuels 1(3):399-408.

Harrill, H., H.-S. Han, and F. Pan. 2009. Combining slash bundling with in-woods grinding operations. Lake Tahoe Council on Forest Engineering. The 2009 COFE Annual Meeting, June 15-18, 2009, Lake Tahoe, California. 14 pp.

Miyata, E. 1980. Determining fixed and operating costs of logging equipment. General Technical Report NC-55. USDA Forest Service, North Central Forest Experiment Station, St. Paul, Minnesota.

Olsen, E., M. M. Hossain, and M. E. Miller. 1998. Statistical comparison of methods used in harvesting work studies. Research Contribution 23. Forest Research Laboratory, Oregon State University, Corvallis.

Pan, F., H.-S. Han, L. R. Johnson, and W. J. Elliot. 2008. Production and cost of harvesting, processing, and transporting small-diameter (<5 inches) trees for energy. Forest Prod. J. 58(5):47-53.

R Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, http:// www.R-project.org/.

Rummer, B. 2008. Assessing the cost of fuel reduction treatments: A critical review. Forest Policy Econ. 10:355-362.

Schatz Energy Research Center (SERC). 2013. RePower Humboldt: A Strategic Plan for Renewable Energy Security and Prosperity. SERC, Areata, California.

Williams, R. B., B. M. Jenkins, and S. Kaffka. 2007. An assessment of biomass resources in California. Prepared for the California Energy Commission. Contract No. 500-01-016. University of California, Davis.

Case Study of Three High-Performing Contract Loggers with Distinct

The authors are, respectively, Graduate Research Associate, Dept, of Forestry and Wildland Resources, Humboldt State Univ., Areata, California (jab22@humboldt.edu [corresponding author]); Assistant Professor, Dept, of Forestry and Landscape Architecture, Korea National College of Agric. and Fisheries, Hwaseong-Si, Gyeonggi-Do, Republic of Korea (hsk5311@Korea.kr); and Professor, Dept, of Forestry and Wildland Resources, Humboldt State Univ., Areata, California (hh30@humboldt.edu). This paper was received for publication in July 2014. Article no. 14-00071.

[c] Forest Products Society 2016.

doi: 10.13073/FPJ-D-14-00071

Table 1.--Hourly cost for the machines used in the centralized
biomass recovery operation.

                                        Initial price   Utilization
Machine                                    ($) (a)      rate (%) (b)

Linkbelt 3400 loader with dump truck       345,000           80
Volvo A35C dump truck                       75,000           80
Linkbelt 3400 loader with grinder          345,000           85
Peterson Pacific 5710C grinder             650,000           85
AWD tractor (d)                             80,000           50
42-ft trailer                                6,000           35
Highway truck tractor                      144,000           85

Machine                                 Total hourly cost ($/PMH) (c)

Linkbelt 3400 loader with dump truck                244.57
Volvo A35C dump truck                               115.59
Linkbelt 3400 loader with grinder                   231.19
Peterson Pacific 5710C grinder                      448.02
AWD tractor (d)                                     117.62
42-ft trailer                                         3.89
Highway truck tractor                                86.03

(a) Modification cost included.

(b) Rates estimated by contractor.

(c) Labor cost included. PMH = productive machine hour.

(d) AWD = all-wheel drive.

Table 2.--Delay-free average cycle time equation (model P <
0.05, [r.sup.2] = 0.1273, n = 482) for loading loose logging slash or
hardwood whole trees into a modified dump truck.

                                Estimate
Parameter                         (s)       SE      F     Mean     P

Intercept                         23.28    3.10   10.05          <0.05
Loaded swing arc                 +0.018    0.01           146    <0.05
  (45[degrees]-270[degrees])
Empty swing arc                  +0.006    0.01           145     0.22
  (45[degrees]-270[degrees])
Pile type (whole tree)            +2.63    1.82                   0.02
Pile type (slash) (a)             +0.00    3.10                  <0.05
Pile location (in unit)           -3.94    3.10                  <0.05
Pile location (roadside) (a)      +0.00    1.86                  <0.05
Pile location (scattered)         +2.44    2.37                   0.09
Pile species (conifer) (a)        +0.00    3.10                  <0.05
Pile species (hardwood)           +3.08    2.20                   0.02
Pile species (mix)                -0.26    1.77                   0.81

(a) Reference variables.

Table 3--Delay-free average cycle time equation (model P < 0.05,
[r.sup.2] = 0.78, n = 62) for a dump truck delivering slash, whole
trees, or mixed loads to a centralized grinding site.

Parameter                            Estimate (s)     SE       F

Intercept                               396.25       90.72   19.63
Traveling empty, distance (ft)           +0.07        0.16
Traveling empty, slope (%)              +46.96      143.37
Positioning empty, distance (ft)         +0.13        0.16
Traveling loaded, distance (ft)         +0.012        0.16
Traveling loaded, slope (%)             +30.33      144.39
Positioning loaded, distance (ft)        -0.08        0.46
Load (slash)                            -82.79      138.63
Load (whole tree)                        +6.46       90.72
Load (mixed) (b)                         +0.00       54.65

Parameter                            Avg. values observed     P

Intercept                                                   <0.05
Traveling empty, distance (ft)               2,325           0.44
Traveling empty, slope (%)                      -1           0.59
Positioning empty, distance (ft)               297           0.17
Traveling loaded, distance (ft)              2,363           0.90
Traveling loaded, slope (%)                      1           0.78
Positioning loaded, distance (ft)              112           0.77
Load (slash)                                NA (a)           0.33
Load (whole tree)                               NA          <0.05
Load (mixed) (b)                                NA           0.84

(a) NA = not applicable.

(b) Mixed material is a combination of whole trees and slash.

Table 4.--Reduced regression model to predict average cycle
time (model P <0.05, r2 = 0.78, n = 70) for dump truck to
deliver material over varying distances.

Parameter         Estimate (s)    SE       F      Mean      P

Intercept            329.76      38.55   248.90           <0.05
Total distance        +0.05       0.01            2,523   <0.05
  traveled (ft)

Table 5.--Productivity and cost of a "hot operation" centralized
biomass recovery system with decoupled highway transportation.

                                             Delay-free
                                            productivity     No. of
Machine                                     (BDT/PMH) (a)   machines

Loader in unit with dump truck                46.57           1
Dump truck                                    26.16           2
Loader with grinder                           38.04           1
Grinder                                       38.04           1
AWD truck tractor (c)                         26.47           2
Decoupled transportation using a trailer
  transfer location Highway truck
  tractor (d)                                 11.92           1-3
Total system                                  38.04 (e)       8-10

                                              Cost      Avg. distance
Machine                                      (S/BDT)      (mi) (b)

Loader in unit with dump truck               6.43            --
Dump truck                                   6.08            0.48
Loader with grinder                          6.10            --
Grinder                                     11.78            --
AWD truck tractor (c)                        6.38            2.13
Decoupled transportation using a trailer
  transfer location Highway truck
  tractor (d)                                7.53           13.00
Total system                                44.30 (f)

(a) BDT/PMH = bone dry ton per productive machine hour.
(b) One-way distance.

(c) All-wheel drive truck tractor and trailer.
(d) Including two trailers.

(e) 38-04 BDT/PMH was the limiting production rate in the system.
Therefore, it was used to calculate cost for each machine in "hot"
system. (Machine cost X number of machines)/limiting productivity
rate.

(f) Total cost does not include other costs such as supporting
equipment (fuel, water, maintenance trucks, overhead, and risk/
profit allowance).

Table 6.--Delays observed for machines used in the collection and
comminution segments of a centralized biomass recovery operation.

                       Total time          Total delay
Machine(s)           observed (min)   observed, min (%) (a)

Loader                    227                 1 (<1)
Dump truck                952               164 (17)
Loader and grinder        139                51 (36)

                     Delay by type,
Machine(s)            min (%) (b)                Reason

Loader                   1 (100)      Operational
Dump truck              71 (43)       Waiting on loader in unit
                        83 (51)       Waiting on second dump truck
                         2 (1)        Waiting on grinder loader
                         7 (4)        Operational
                         1 (1)        Personal
Loader and grinder      36 (71)       Waiting on dump truck
                         5 (10)       Remove choker
                         6 (11)       Operational
                         4 (8)        Personal

(a) Percent delay of total time observed.

(b) Percent delay of total delay observed.
COPYRIGHT 2016 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bisson, Joel A.; Han, Sang-Kyun; Han, Han-Sup
Publication:Forest Products Journal
Geographic Code:1U9CA
Date:Jan 1, 2016
Words:6688
Previous Article:Carbon mitigation impacts of increased softwood lumber and structural panel use for nonresidential construction in the United States.
Next Article:Case study of three high-performing contract loggers with distinct harvest systems: are they thriving, striving, or just surviving?
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters