The Cost-Efficiency of Seedling Packaging Specifically Designed for Tree Planting Machines

Today’s crane-mounted planting heads plant seedlings with biologically similar or better results than operational manual planting. However, the total cost of mechanized tree planting in southern Sweden must decrease at least 25% to compete economically with manual planting. Although seedlings packed in machine-specific packaging increase the productivity of planting machines by reducing seedling reloading time, they also increase logistics and investment costs. In this study, we analyzed the total cost of outplanting seedlings with an excavator-mounted Bracke Planter and seedlings packed according to four different concepts: cultivation trays, cardboard boxes, band-mounted seedlings in cardboard boxes and linked pots in container modules. The total cost per planted seedling was calculated for each packaging system as the sum of all costs from nursery to the recovery of empty packaging. The results showed that today’s system of transporting seedlings in cultivation trays is the most costefficient of the four alternatives. Machine-specific seedling packaging was 16–23% costlier per planted seedling than cultivation trays when trucking distances were 100 km. Sensitivity analyses indicated that machine-specific seedling packaging increased in cost-efficiency relative to cultivation trays primarily when more planting machines were contracted, but also as planting machine fixed costs and productivity increased. Moreover, the relative cost-efficiency of band-mounted seedlings, but not seedlings in container modules, increased with increasing trucking distance. Thus, we show that investments in machine-specific seedling packaging for today’s planting machines are justified only when the fixed costs, productivity and number of contracted planting machines increase substantially.


Introduction
In the Nordic countries, stand regeneration after clearcutting is predominantly performed using manual tree planting.As an alternative to manual tree planting, mechanized tree planting has been revived in southern Sweden in the last five years.In contrast to the high investment, highly productive, continuously operating planting machines of the 1980s-90s (Hallonborg 1997), today's machines use low investment, intermittently advancing, crane-mounted planting heads like the Bracke Planter, EcoPlanter (Hallonborg et al. 1997, Saarinen 2006) and M-Planter (Rantala et al. 2009).Currently, however, mechanized planting in southern Sweden is only performed using the Bracke Planter and at a higher cost than manual planting.
One of the key reasons why crane-mounted planting machines are still not economically competitive with manual planting in Sweden is the lack of automated seedling handling and feeding systems.The planting heads must instead be reloaded manually, and this task can occupy from 15% (Rantala et al. 2009) to more than 30% of the machine's effective working time (Halonen 2002).Manual reloading entails that the planting machine's capital costs and driver's wage are still present while the machine's production is zero.Moreover, as machine productivity increases, relatively more of the effective working time is spent manually reloading seedlings (Halonen 2002).There is, therefore, a need to automate the feeding and handling of seedlings if crane-mounted planting machines are to become cost competitive compared to manual planting (Normark andNorr 2002, Sönsteby andKohmann 2003).
Historically, there have been few attempts to automate the seedling feeding and handling systems on tree planting machines, none of which are in operation today.For example, the Serlachius planting machine, developed in the late 1970s, used a chain-driven conveyor to feed seedlings automatically to the planting arms from either Hiko growth containers or special peat pots that were sawn apart (Stjernberg 1985, Hallonborg 1997).The latter were part of an integrated cultivation system including specially designed pallets on which 384 seedlings per pallet could be cultivated, transported and used as the basic unit for the automatic feeding system (Kohonen 1981).
In the mid 1990s, an attempt to automate the seedling feeding system on the Silva Nova planting machine was also made (Hallonborg 1997).This feeding system was called the Pot Link System (PLS) and used plastic pots linked together to form belts.The PLS-belts were loaded at the landing by a portable loading robot using standard seedlings delivered from the nursery in their growth containers, and then automatically shunted into a large container holding approximately 7000 seedlings.Finally, the large container was placed on the Silva Nova using the planting machine's crane.
A few years later, a concept called EcoBandPak (EBP) was developed for the EcoPlanter (Normark and Norr 2002).The EBP was based on the band-mounted seedling concept in which seedlings were mounted one after another between two strips of plasticized paper.Thereafter, the coils of band-mounted seedlings were packaged into cardboard boxes, transported to the planting machine and then manually inserted into a prototype automatic feeding system.
The efficiency of automated handling and feeding systems is also dependent on how seedlings are packaged and transported.For example, the planting machine loses productivity if the operator must unpack seedlings from cardboard boxes or move them from nursery trays and then load them into an automatic seedling feeding system (Sönsteby and Kohmann 2003).Considering that today's containerized seedlings in southern Sweden are delivered to the planting machine either in nursery trays or cardboard boxes designed for manual planting, machine-specific seedling packaging solutions are necessary if automatic seedling feeding systems on planting machines are to become cost-efficient.
However, machine-specific packaging solutions warrant extra labour, packaging, and investment costs at the nursery and extra investment costs for the planting machine (Normark and Norr 2002).Seedling transport costs and work methods might also be affected.In one previous study, Lawyer and Fridley (1979) modelled how seedling transport affected the total cost of mechanical planting systems under North American conditions.In their model, Lawyer and Fridley used Eq. 1 to show that the total cost per hectare of mechanized tree planting systems can be divided into three parts as follows: where C S = nursery costs; C T = cost of transporting seedlings from the nursery to the planting site; C F = cost of planting machine.Using Eq. 1 as a starting point, this present study investigates if specifically designed seedling packaging systems which raise C T can lower C F enough to reduce the total cost of mechanized planting (C Total ).Accordingly, the aims of this article are twofold: 1) to describe two seedling packaging concepts specifically designed for planting machines; and 2) to compare the total cost of these two packaging concepts with today's two most common containerized seedling packaging systems used during manual planting in southern Sweden.

Materials and Methods
This comparative cost analysis is based on data (costs and time requirements) collected from nurseries, contractors and relevant companies during winter 2010.To correct for inflation biases, all costs were adjusted to December 2009 values according to Statistics Sweden's (SCB) Consumer Price Index.The average exchange rate for Swedish kronor (SEK) to Euros (€) from January 2009 to December 2010 was SEK 10.08/€ (Sveriges Riksbank 2010).
Most time consumption data were acquired by interviewing machine operators and nursery personnel; however, time studies of some crucial tasks in the transportation chain were used to complement the interview-data.When calculating depreciation costs, the reducing balance method was used (Rantala et al. 2009).

System Delimitations and Models
The four seedling packaging systems (s1-s4) used in this cost analysis are described below.Seedlings in Hiko cultivation trays are the starting point for all systems, but with additional seedling packaging in s2-s4.The packaging systems were expressed as models based on the activities of the generic transportation chain shown in Fig. 1.The chain starts at the nursery with the seedlings still being in their cultivation trays while aggregated on large frames after having been sorted and sprayed with insecticides.The chain ends after outplanting when the empty seedling packaging has either been returned to the nursery for reuse or recycled.
The cost analysis was based on Eq. 1 from Lawyer and Fridley (1979) but modified so that C S was defined as all costs of cultivating seedlings up until packaging them for transport.This kept C S constant and was therefore not included in this study's model.Thus, the total cost per planted seedling of mechanized planting (C Total , expressed as SEK per planted seedling (SEK/ pl)) with s1 was calculated using Eq. 2, while s2 and s3 were calculated using Eq. 3, and s4 with Eq. 4.  Unless otherwise stated, all activity costs were expressed in SEK per seedling (SEK/pl) and calculated according to Eq. 5 where c is the hourly cost for activity i and t is the time consumed for the performed work (h/pl).

Nursery Investments and Packaging
The seedlings in the analysis were 1.5 year old Picea abies (L.) Karst grown in copper-painted Hiko v93 trays for circa 20 months at a midsized nursery (annual production of ten million seedlings) in southern Sweden.These seedlings were assumed to be sorted into a standard height class of 20-30 cm stem length and sprayed with insecticide against Hylobius abietis (L.) while in Hiko trays three days before being packed and shipped.s1 and s4 assumed hot-lifted seedlings which require daily watering while s2 and s3 assumed frozen-stored seedlings which do not require watering the initial three days after thawing.The outplanting window, altogether 190 days, was assumed to be April 1 to November 15 but with no planting during July.
The costs of the four systems' packaging materials are shown in Table 1.The cost of s1 reflected the extra damage and wear inflicted on the Hiko cultivation trays by using the trays as seedling packaging.The interest rate was 10% for all investments by the nursery company.
Besides the material costs, C Pack included the extra costs of all machinery and nursery activities needed to package the seedlings.These extra costs were absent from s1.For s2, these extra costs included the use of a small forwarder to shuttle multiple frames of seedlings from the pesticide quarantine/spraying area to a packing line; the investment and operative costs of a fully automatic packing line (including box erector, packing machine, palletizer, and stretch wrap machine) housed within an insulated production hall; and the use of a small front-end loader to move loaded pallets to a covered storage area.s2 was assumed to pack the nursery's entire yearly production due to cardboard boxes also being used for manual tree planting.
For s3, the extra costs included the use of a small front-end loader rather than a forwarder; the investment and operative costs of a manually operated band-mounting line (in which seedlings were manually band-mounted and boxes were manually erected, packed, sealed, and stacked); an automatic stretch wrap machine (all machines housed within a insulated production hall); and the use of a small front-end loader to move loaded pallets to a covered storage area.However, when the yearly number of seedlings packed in s3 surpassed 1.2 million, it was assumed that s3 entailed further investments in an automatic box erector, one additional band-mounting station and a plantgripping robot (for automatic band-mounting of seedlings).
For s4, the extra costs included the use of a small front-end loader to shuttle multiple frames of seedlings from the pesticide quarantine/spraying area to the packing line; the investment and operative costs of a transplanting machine (for moving seedlings from Hiko trays to PLS pots, Total material cost per seedling SEK/pl a) 0.12 0.10 0.17 0.07 a) = seedling b) Maximum weight of regularly manually handled units according to workplace ordinances (Lumsden 2006)  c) According to space requirements for upright-standing seedlings in Österström et al. (1974)  d) Includes cost of paper band and loading PLS trays into the containers) housed within an insulated production hall; and the use of a small front-end loader to move containers to and from the packing line.
Table 2 summarizes the cost data for the s2, s3, and s4 packaging costs.The depreciation period of the packing machines and production halls were 10 and 25 years respectively, the salvage values were 10% of the purchase price, and the yearly insurance costs were 0.15% of the assets' value.At the nursery, the hourly cost of labour including administration and taxes was 210 SEK, of a small front-end loader including driver was 270 SEK, and of a small forwarder including driver was 435 SEK.The total construction cost of a new production hall was 3500 SEK per m 2 .

Seedling Transport
For s1, seedling trays were assumed to be manually loaded at the nursery by the courier truck's driver, two nursery workers, and one small front-end loader.The Hiko trays were placed lying sideways, facing each other, with the seedlings' stems intertwined so that the distance between the two trays equalled 1.5 times the average length of one seedling's stem (Österström et al. 1974).500 cultivation trays was the space limit of the 3 ton courier truck, and this limit also meant that no seedlings were stored at the contractor's depot for longer than two weeks.At the contractor's depot, s1 entailed the manual unloading by both the courier truck's driver and the planting machine operator.Because the courier truck was assumed to be solely commissioned for seedling transport, transport costs for s1 also included the cost of the courier truck's return trip.
For the other systems, pallets and containers were loaded by the nursery's small front-end loader onto general groupage delivery trucks.For s2 and s3, unloading at the contractor's depot was handled by the delivery truck driver alone using the truck's tail lift and hand pallet truck.For s4, the delivery truck driver used a truck-mounted forklift during unloading.Because of their excess  a) Based on empirical data from nurseries and a first-generation packaging machine b) Based on empirical data from the prototype EcoBandPak packaging machine height, shipping units in s3 and s4 were assumed to not be stacked double within the delivery truck as was done with s2's pallets.Because of desiccation concerns, frozen-stored seedlings could not be stored at the contractor's depot for longer than three days.This rule limited each truck load for s2 and s3 to three days worth of seedlings.Meanwhile, for s4, balancing container module investment costs against scale economies during primary transport meant that five shipping units per truck load was the most cost-efficient.
At the contractor's depot, s1 and s4 included the cost of daily watering of stored seedlings by the planting machine operator.For s2 and s3, this cost was instead replaced by the cost of freezer storage (Table 2).
Seedling transport from the contractor's depot to the landing was assumed to coincide with the planting machine operator's daily travels to and from the clearcut.Therefore, the cost of secondary transport referred only to the time needed for the planting machine operator to load and unload seedlings from his personal vehicle.Since this workplace time would otherwise be effective planting machine work time, all tasks performed by the planting machine operator were assumed equal to the planting machine's fixed costs including operator wage.For s4, however, secondary transport entailed an investment in a pickup truck equipped with a small manuallyoperated hydraulic crane allowing for loading and unloading containers.Nevertheless, the variable costs (e.g.fuel and maintenance costs) of s4's secondary transport were assumed to equal the other three systems' variable costs.The depreciation period of secondary transport vehicles was five years for all systems.The interest rate was 6% for all investments by the contractor.Table 3 summarizes the transport costs of the four packaging systems.

Outplanting
The cost analysis assumed seedling supply to two planting machines, and that these were both Bracke Planters mounted on a used 16 ton Kobelco 135BSR excavator; that the yearly use of the planting heads and the base machines was 1000 hours and 1100 hours respectively (where the base machine works 100 hours per year with other duties); and that 200 000 seedlings per year per machine were planted using a single-shift work system.However, when productivity rose, more seedlings were assumed to be planted per year instead of the machine only working fewer hours per year (and vice-versa for lower productivity).Moreover, in this analysis, the planting machine's annual relocation costs were considered constant regardless of machine productivity.Table 4 lists the main factors assumed when calculating the cost of the planting machine.
For s1, s2, and s3, seedlings were assumed to be loaded from the operator's vehicle directly into the planting machine's storage box.Thereafter, the planting head was reloaded using two handling units for s1 and one handling unit for s2 and s3.For s3, reloading included removing the seedling coil from the cardboard box and positioning the band into the feed rollers.For s4, containers were assumed to be offloaded from the pickup truck before being lifted onto the side of the base machine by a hydraulic lift.Thereafter, the planting head was reloaded with 200 seedlings using the PLS pots formed into two separate trays of 100 seedlings each.These trays were removed one at a time from the container and then hooked together on the planting head to form one long chain of pots.During planting, seedlings were automatically extracted from this chain and fed into the planting tube.
Empty packaging was assumed to be organized and placed back into the storage box during each reloading occasion.The time required for these tasks was included in the reloading cost.For s4, the empty PLS chain was assumed to be broken up into two separate trays by the operator before being placed back into the container.
As can be seen in Table 5, the packaging systems other than s1 also required modifications to the seedling storage box or planting head.These investments were assumed to have no salvage value and a depreciation period of 5 years.

Returns and Recycling
All empty packaging was returned to the contractor's depot after each shift.From there, empty Hiko trays in s1 were returned two times per year to the nursery by the returning courier truck.Unloading empty Hiko trays at the nursery assumed the help of one nursery worker and a small front-end loader.For s2 and s3, flattened boxes were stacked and tied onto empty Euro pallets that were picked up once a year by a recycling company.For s4, empty containers were transported back to the nursery by the general groupage delivery trucks in conjunction with each delivery of loaded containers to the contractor's depot.Unloading empty containers at the nursery assumed the help of a small front-end loader.
For s4, containers were rinsed out by a nursery worker using a water hose after each delivery.This task was assumed to take 25 minutes.Moreover, the PLS pots were washed once a year at the end of the season in the nursery's standard tray washing machine to avoid peat residue build up.This washing task was assumed to cost 0.05 SEK/ pl plus require an additional 30 minutes of labour per container module.For s1, however, the cost of washing Hiko trays was not included because trays were washed before being used for cultivation regardless of the tray having left the nursery   of not.In Table 6, handling time per shift includes all necessary tasks when handling empty packaging from the planting machine to the depot.

Sensitivity Analysis
The basic scenario assumed used base machines, relatively short primary transport distances, relatively high interest rates and few contracted planting machines (Table 7).Therefore, the sensitivity analysis focused mainly on investigating the impact of higher fixed costs (e.g. through new base machines), longer primary transport distances, lower interest rates and more contracted planting machines on the cost analysis.Table 7 gives the sensitivity analysis' factor variation range.Mean planting time excluding seedling handling is a measure of the planting machine's overall productivity where -40% reaches the theoretical limit of the Bracke Planter.
Planting machine-specific seedling packaging is also relevant for the M-Planter, a two-headed, crane-mounted planting head.According to Rantala et al. (2009), this type of planting machine increases planting productivity per productive hour (E 15 , delays shorter than 15 minutes included) by 36% while only increasing total costs by 4.1% compared to the Bracke Planter.Therefore, a supplemental analysis was done using the M-Planter instead of the Bracke planting machine in which reloading the M-Planter planting head for s1, s2, s3 and s4 was assumed to take 366, 492, 420 and 460 s respectively.

Basic Scenario and Sensitivity Analysis
Under the basic assumptions, s1 (the reference system) was the most cost-efficient packaging system; it was 12.0%, 16.4% and 23.1% less costly than s2, s3 and s4 respectively (Table 9).Both s3 and s4 were burdened by high nursery investment costs when only two planting machines were contracted.Consequently, the  a) Loading cost of the groupage delivery truck is included within the total transport cost for returns  interest rate affected the relative cost-efficiency of the packaging systems to a large extent although s4 benefitted relatively most from a lower interest rate (Fig. 2F).
Although the values of C Total all decreased, the packaging systems' relative order of cost-efficiency did not change when using the M-Planter compared to the Bracke Planter in the cost analysis.s1 was still ca 10% to 18% more cost-efficient than the other packaging systems.As can be seen in Fig. 3, varying the relative planting head reloading time had little effect on this result.

System Investment Costs
The total supplementary investment cost of s1, s2, s3 and s4 in the basic scenario was 0.12, 0.29, 0.82 and 0.88 SEK/pl respectively.s1 entailed no additional investments for the contractor, while 4%, 8% and 36% of the total investment cost was borne by the contractor for s2, s3 and s4 respectively.s4 resulted in the largest total investment cost per seedling except when there were precisely five contracted planting machines.Five planting machines lead to sufficient demand for band-mounted seedlings to warrant an investment in an automatic bandmounting line for s3.This additional investment momentarily raised s3's total investment cost per seedling above that of s4.higher cost of s3 compared to s2 was mostly attributed to the low capacity utilization rate of the nursery's band-mounting machine.Similarly, s4 suffered from the low capacity utilization rate of the nursery's transplanting machine as well as high costs for handling, transporting and washing empty container modules.
As revealed by Fig. 2A, s1 was most costefficient even when planting head reload time was decreased by 50%.Halving the reload time increased total planting machine productivity the least for s3 (3.9%) because its baseline reloading time was relatively low to begin with.In contrast, s3 benefitted the most from increased planting machine fixed costs (Fig. 2B); increasing the fixed costs by 50% made s3 slightly more cost-efficient than s2 (0.6%).Primary transport distances longer than 100 km affected s1 most negatively (Fig. 2C) because the cost benefit of the other systems' groupage trucks were first realized over these longer distances.Shorter distances, on the other hand, affected s4 the most which reflected s4's added cost for a truck-mounted forklift.Likewise, s4 was the most sensitive to changes in the mean planting time excluding seedling handling (Fig. 2D).s3 was slightly more cost-efficient than s2 when mean planting time decreased by 30%, and s4 was more cost-efficient than both of them when the planting time was 40% lower.Neither the mechanical availability (MA) of the planting head's seedling feeding system (Fig. 2E) nor  Because s1 and s2 entailed no planting machinespecific investments for the nursery, an increased number of contracted planting machines only reduced the total cost of s3 and s4 (Fig. 4).Therefore, s3 became more cost-efficient than s2 already with five contracted planting machines.With 20 planting machines, s3 became only 3.4% costlier than s1 while s4 became 0.6% more cost-efficient than s2.The nursery's supplementary investment cost for s3 and s4 fell from respectively 0.75 and 0.56 SEK/pl in the basic scenario to 0.34 and 0.18 SEK/pl with 20 planting machines.

Best and Worst Case Scenario
The primary transport's effect on the cost-efficiency of planting machine-specific packaging systems s3 and s4 in the best case (35% higher planting machine fixed costs, 20% lower mean planting time, 3% interest rate, and 20 contracted planting machines) and worst case scenario (same fixed costs, 10% higher mean planting time, 12% interest rate, and two contracted planting machines) is shown in Fig. 5.In these scenarios, the MA of the planting head's seedling feeding system for s1 and s2 was assumed to be 100% while it was lowered for both s3 and s4 to 97% and 85% in the best and worst case respectively.
The difference in cost-efficiency for s3 between the best case and worst case scenario at 100 km primary transport distance was 37%, whereas s1 and s2 only varied 3.2% and 2.4% respectively.Nonetheless, s3 was, despite being penalized with a lower MA compared to s1 and s2, only 0.9% more expensive than s1 in the best case scenario at 100 km primary transport distances, while more cost-efficient at primary transport distances of 200 km and longer (Fig. 5A).

Discussion
According to the proposed model, s1 was more cost-efficient than the other three packaging systems under southern Swedish conditions.However, decreasing the mean planting time increased s3 and s4's cost-efficiency, albeit to a small extent.As can be seen in Fig. 2D, the negative slope of s3 and s4 are steeper than s1 and s2.This difference in slope indicates that as planting machine productivity increases, the fixed costs of s3 and s4's additional investments decrease and these packaging systems become more cost-competitive relative to s1 and s2.Similarly, Fig. 2B illustrates the connection between higher planting machine fi xed costs and increased cost-effi ciency of the planting machinespecifi c packaging systems.Because s1 and s2's slopes are steeper in this graph, their cost-efficiency decreases more rapidly than s3 and s4 when fi xed costs increase.Thus, increasing planting machine fi xed costs in the basic scenario by 35% to refl ect the cost of a new Kobelco 135BSR excavator and Bracke Planter made s3 10.5% and 0.6% costlier than s1 and s2 respectively.
The model assumed that seedlings in s3 were band-mounted by custom-built nursery machines based on a prototype developed in the 1990s.There is probably improvement potential for such prototype machinery.If the productivity of the manual band-mounting station could double from 1000 to 2000 seedlings per hour, C Total of s3 in the basic scenario would decrease by 0.15 SEK/pl, or 3.9 %.This adjustment in productivity would be enough to make s3 0.3% more cost-effi cient than s2.
According to time studies, s3 and s4 could decrease planting head reloading time compared to s1 by 61% and 55% respectively.At the same time, s3 and s4 invariably make the planting head more complex by adding parts and creating series-parts reliability dilemmas (Bowen 1981).
The basic scenario assumed no difference in the planting head's mechanical availability (MA) between the four packaging systems.This might be unrealistic, however, since the added complexity of s3 and s4 would probably lower the MA of planting heads using these packaging concepts (cf.Mellgren 1989).This added complexity is the reason why only s3 and s4's MA was reduced in the best and worst case scenarios.
The proposed model is a comprehensive theoretical cost-analysis without the effects of operators, site characteristics and different technical maturity levels which are otherwise present in empirical studies.The absence of such disturbances makes the model robust.However, since the input values are time-and-place specifi c, the total costs presented here might be less relevant than the systems' relative cost-effi ciency and the factors which infl uence this correlation.Nonetheless, by changing the input values, this model should be widely applicable in the boreal context and not just limited to southern Sweden.
Because seedlings in s1 and s4 must be hotlifted, they have actively growing shoots in May and June (Luoranen and Viiri 2005).Actively growing shoots are fragile and cannot be graded by an automatic grading line.Thus, s1 and s4 must in reality include the additional cost of manual grading at the nursery.For s2 and s3, however, seedlings can be graded and packed during winter months when the seedlings are dormant.Packaging in winter also has the advantages of handling seedlings when they are most stress resistant (Ritchie and Landis 2010) and spreading out nursery work over the whole year rather than concentrating all tasks to the growing season.
Moreover, dormant seedlings respire less and thus dry out more slowly than hot-lifted seedlings.This means that despite watering, seedlings in s1 and s4 are more likely to be planted in a less vital state compared to s2 and s3.Similarly, seedlings in s1 and s2 are exposed to detrimental abrasion after being loaded onto the carousel on today's Bracke planting head.This abrasion can cause the root plugs of seedlings grown in copper-painted cultivation trays to fall apart, thereby causing root damage and deformation during planting.Seedlings mounted in paper bands (s3) or carried in PLS pots (s4) would be protected from this abrasion.Even if these potential differences in seedling vitality were not accounted for in the proposed model, it is important to identify the packaging systems' nonmonetary benefits.
C Total of s3 decreased relatively less than the other packaging systems when using the M-Planter compared to the Bracke Planter in the cost analysis (Fig. 3).s3 and s4's lack of costefficiency with the M-Planter can be attributed to this planting head's faster reload time.Rantala et al. (2009) measured 25.8% less time per seedling when reloading the M-Planter planting head with seedlings from cultivation trays compared to the Bracke Planter, and this lower time consumption limits s3 and s4's window of opportunity for productivity improvement.
Compared to the planting machines available today, even greater productivity improvements during outplanting could be expected if an intermittently working machine could use more planting heads, for instance four.This concept would entail a much more expensive and productive machine because it is probable that it would need two cranes.This base machine would have to be custom-built which would vastly increase the planting machine's fixed costs.These higher fixed costs would allow more economical leverage for planting machine-specific packaging concepts like s3 and s4.
Today, the Bracke planting machine plants seedlings with biologically similar (von Hofsten 1997, Saarinen 2006) or better (Ersson and Petersson 2009) results than operational manual planting.However, the total cost of today's Bracke Planter must decrease by at least 25% if it is to be economically competitive compared to manual planting in southern Sweden.As mentioned above, a two-craned planting machine using four planting heads and machine-specific seedling packaging would probably increase productivity enough to lower the total cost of mechanized planting by 25%.Nevertheless, such a machine would require some degree of semi-automation (shared control) to enable the operator to control two cranes simultaneously.Simulation studies are needed to verify the potential productivity increase of such a two-craned planting machine.

Conclusion
Under the circumstance of only contracting two relatively inexpensive and unproductive planting machines, no investment in machine-specific packaging systems is justified.However, an increase in the fixed cost, the productivity or especially the number of contracted planting machines increases the cost-efficiency of such packaging systems.

C
PM.Load = cost of loading seedlings onto planting machine, C Outplant = total cost of outplanting seedlings including reloading planting head, C Returns = cost of returning multi-use packaging to nursery, and C Total.Trans = total cost of transporting seedlings with groupage delivery trucks including loading and unloading, C Recycling = cost of recycling single-use packaging (cardboard boxes).

Fig. 2 .
Fig. 2. The effect on the total cost of the four seedling packaging systems when varying (A-F): the relative planting head reload time; the planting machine's fi xed costs; the primary transport distance; the mean planting time excluding seedling handling; the planting machine's mechanical availability; and the interest rate.

Fig. 3 .
Fig. 3. Total cost of the four packaging systems with two M-Planter planting machines.

Fig. 4 .
Fig. 4. Total cost of the four packaging systems when contracting more planting machines.

Table 1 .
Cost factors for packaging materials.

Table 2 .
Cost factors for packaging machinery and extra nursery activities (based on expert assessments unless otherwise stated).

Table 3 .
Cost factors for primary (from nursery to contractor's depot) and secondary (from depot to landing) seedling transport and watering.
c) Loading and unloading costs of the groupage delivery truck is included within the total primary transport cost d) Unloading trays and boxes during secondary transport is assumed to coincide directly with loading seedlings into the planting machine's storage box

Table 4 .
Baseline cost factors for the Bracke planting head and base machine.

Table 5 .
Cost factors for storing seedlings on planting machine and reloading planting head.

Table 6 .
Cost factors for handling empty seedling packaging.

Table 7 .
Change in factors from the basic scenario.

Table 8 .
Cost factors for the M-Planter planting machine.
a)Including delays shorter than 15 minutes

Table 9 .
Aggregated costs in the basic scenario for the four seedling packaging systems.