# Feasibility Study of Digital Manufacturing Systems Applied for Medium Scale Production Marlon Wesley Machado Cunico Abstract-Purpose: Along the last years, the complexity of products has been growing progressively, while the product development life-cycle tended to be reduced. In addition to that, additive manufacturing technologies increased their role in the product development process, resulting in reduction of errors and products release time. In spite of these benefits, the main application of these technologies is still focused on initial phases of projects and results in high costs of parts and low volumes. On the other hand, although conventional produtivity processes results in low costs and high volumes, the investiment related to these processes are high and the implementation time are long. For that reason, the main goal of this work is to investigate the possibility of application of additive manufacturing technologies for small and medium scale production. Design/methodology/approach: In order to analyse the feasibility of additive manufacturing technologies as productivity way, we established the injection molding as the reference process. In addition, it was also studied 4 scenarios where the volume of parts, time demand, parts maximum dimension and flexibility were the variables and the lead time, part cost, investments cost and pay-back period were the responses. In the first scenario, it was analysed the production feasibility of a injection molding process where the injector machine and tooling costs were considering as capital investment. In the second scenario, we investigated the feasibility of additive manufacturing services for production, while the forth and fifth scenarios analysed two different production strategies where additive manufacturing technologies are considered. In both cases, the acquisition of equipment was considered in the investment and part cost estimation. At the end, all the scenarios were compared in order to identify suitability the production strategy for small and medium scale strategies. Findings: As result of this study, it was possible to identify part cost estimation models for different sort of production ways, where the feasibility of this scenarios could be evaluated. It was also seen the variation of part cost as a function of annual demand in addition to the analysis of minimal stock analysis, lead time and demand time. Through these analyses, it is possible to identify the feasibility of each one of studied production ways in accordance with annual part demand. By the end, all the studied scenarios were compared and it was possible to indicate the most suitable production way as a function of annual part demand. In this case, very small production scale was marked to be better attended by additive manufacturing services, while small scale was by low cost additive manufacturing with 8 machines in network arrangement. ver the last several years, the application of additive manufacturing (AM) processes has been steadily growing up as consequence of the advantages provided by it sort of process. In contrast with the benefits of these technologies, the main application is still focused on prototypes and special parts, as such medical devices (GIBSON et al., 2002;GIBSON et al., 2010; CUNICO e CARVALHO, 2013b; a). In parallel to those facts, the current production strategies are based on rapid or definitive tooling, resulting in high capital investments. As consequence, small scale production investments tend not to be justified or the payback time is too long (RUFFO et al., 2006). For that reason, the main goal of this work is to present a proposal of small scale production which is based on additive manufacture technologies. As result, it is expected that the analysis and comparison of the current manufacturing process (injection molding) versus 3 other additive manufacturing options indicates the solution that is more suitable for small scale production. In order to analyse the feasibility of additive manufacturing technologies as an effective production way, we established the injection molding as the reference process in addition to studying 4 scenarios where the annual part demand, time demand, parts size and investment were the variables and the lead time, part cost, investments cost and pay-back period were the responses. In all the studied scenarios, we defined and indicated numerical models for the part cost estimation, where the definition of the main components of cost, lead time and minimal stock help to identify the O Introduction feasibility of each scenario according to the part demand. In the first scenario, it was analysed the production feasibility of an injection molding process where the injector machine and tooling costs were considering as amortised capital investment. In this scenario, besides the analysis of feasibility, we have also presented an estimation model for the part, tooling and overhead costs, being useful for the process selection and the part cost estimation at the beginning of projects. In the second scenario, we investigated the feasibility of additive manufacturing services for production, where the costs related to production overhead and tooling are ignored. In this scenario, it is also important to see that besides the lead time and inventory dimensioning play a fundamental role in this business segment, these parameters might determine the feasibility of a new product release. In the third and fourth scenarios, we analysed two different production strategies where additive manufacturing technologies are considered. In both cases, the acquisition of equipment was considered in the investment and part cost estimation. At the end, all the scenarios were compared in order to identify suitability the production strategy for small and medium scale strategies. In addition to the feasibility analysis of small scale production products, this work can also be a very useful tool for customised or tailor-made business segments, where the feasibility of new products is hardly achieved and the product cost tend to be extremely high. # II. # Production Estimation Models In order to investigate and compare the part cost of parts which are made in conventional injected mold and additive manufacturing techniques, we selected the main components of the part cost and created estimation cost models of these components. In general way, the main part cost components can be classified in direct and indirect costs, as shown in Figure 1 In this way, it is possible to see that the direct cost is related to the material which is directed used to fabricate the part, while the indirect costs concern the process time, labour, investments and are amortised by the volume of fabricated parts. ). Therefore, the part cost ( part C ) might be defined as: machine overhead tooling direct part C C C C C + + + = (1) In addition to the analysis of part cost, it was also analysed the feasibility of the production scenarios with respect to demand and lead time. In fact, this is an important parameter to be analysed because it indicates whether the productive way is feasible, in addition to indicating the minimal stock which is necessary for each annual demand and the part demand time. For the definition of part demand time, we assumed that the annual demand is distributed homogeneously along the year. Therefore, it was possible to see that the part demand time might be characterised by: ), as shown in Eq. ( 3) ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? = demand down demand lead down delivery t t round N t t round N Stock (3) Therefore, if we assume that the demand time is equal to the delivery time per part, it is possible to estimate the minimal inventory which is needed to attend production through the maximum of stock curve in addition to the safety stock. # a) Injection Molding In order to identify the total cost of an injected part, we applied an estimation cost model to identify the tooling cost, while the machine cost was obtained by quotation. The part direct cost was estimated through the part volume and the raw material cost was identified by low volume quotation. waste material part material part part direct C k V C + ? ? = _ _ ? (4) # ii. Tooling Cost For the estimation of tooling cost, we applied Boothroyd and Dewhurst's method, which concern in the estimation of operational and direct costs which are necessary to build a cold runner mold (REES, 1996 ), as it is possible to see in Eq. (6). In this study, we ignored the discount factor per cavity in addition to establishing that the number of cavities should be limited to 5. This restriction in the number of cavities was defined because the mold was design to low production volumes (REES, 1996; CHATAIN e DOBRACZYNSKI, 1997; FAGADE e KAZMER, 2000). # ( ) discount cavity cavities cavitiy cavities f n C C _ ? ? = (6) i. Direct Cost The direct cost of part is related to the quantity of material which is necessary to fabricate the part, where the volume of part ( part V ), the specific weight of material ( With respect to the cavity set cost, it is possible to identify that the main cost components are related to cavity material ( ). This estimation can be seen in Eq. ( 8 k V C _ _ _ _ ? ? = ? (8) Where: ) are also included in the model, Eq. ( 14). In spite of the effect of this factors effect on the time estimation, we established that the efficiency of machining is high and the complexity of part was low, resulting in those factors being equal to cavitiy cavitiy cavitiy material cavitiy H W L V ? ? = _ (9) And, ] , 1 . 0 max[ part part part cavitiy H L L L ? + = (10) ] , 1 . 0 max[ part part part cavitiy H W W W ? + = (11) ] 2 , 057 . 0 max[ part cavitiy H H ? =(12f f f t t t _ _ _ _ _ ? ? ? ? ? ? ? ? ? ? + = (14) In addition, the estimation of the cavity volume machining time can be identified by the volumetric mold material removal rate, Eq. ( 15 R h H d h V R V t 1 7 . 0 _ _ _ _ _ ? ? ? ? = = (15) Where: F d A R A t 1 5 . 0 _ _ _ _ ? ? = = (16) With respect to mold base cost and customization costs, we defined that these cost are respectively 15% and 150% of cavities cost. As the tooling cost is considered a capital investment, the contribution of tooling for the part cost is amortised by the volume of parts which is fabricated. Therefore, the total tooling cost might be defined as a function of molding tool cost ( For the determination of total batch time, we defined that the main components that contribute for the batch are the direct molding cycle time ( For hot runner mold systems, SRF is equal to 1, while for cool runner systems SRF is characterised by: ( ) 1 5 . 1 5 . 0 + = part Wt SRF (23) It is also possible to identify the melt capacity of machine through: ) ( 30 ) ( ) ( s t g Shot g Melt Cycle capacity capacity ? = (24) Through this parameter, it is possible to identify the clamp force of machine in a preliminary way according to the Figure 2. On the other hand, it is also possible to estimate the machine clamp force through the wall thickness method. In this method, the clamp force ( clamp F ) is found by the projected area of runner and cavities ( projected A ) and a wall thickness factor, which can be seen in the Eq. (25). As consequence, the general estimation of machine cost might be estimated by the update regression of (BOOTHROYD e DEWHURST, 1988): thickness wall projected clamp f A t F _ ) ( ? = (25)) ( 430 16000 t F C clamp equipment ? + = (26) # b) Additive Manufacturing Services For the study of additive manufacturing services as productivity way, we have also considered that the tooling cost, machine cost and production overhead were null, while the direct cost remained the responsible component of part cost. It is important to note that there are other costs inherent to this kind of scenario, as such logistics, stock and quality. Nevertheless, these costs were ignored in this study in order to create a comparison criterion among the studied scenarios (RUFFO et al., 2006). In order to identify the cost estimation of additive manufacturing services, it was requested quotation of 3 dimension parts, 3 technologies, and 4 part quantities in order to be obtained a statistical regression and the cost estimation formulation. The investigation matrix used to analysed the service cost estimation can be seen in Table 2, where it is related the part quantity per order, main dimensions of analysed parts and fabrication technology. # Table 2: Investigation matrix of additive manufacturing services In addition to being identified the regression equation for each technology and the part cost estimation model, it was also determined the minimal stock volume which would be needed to attend annual part demand. Therefore, the feasibility of this productivity concept might be evaluated. It was also found that the lead time was a based on service time that the bureaux provide ( batch t ) # c) Additive Manufacturing Production For the generalised cost estimation of additive manufacturing parts, it was analysed the direct cost and production overhead as a function of part size, building area, parts demand and batch volume. Therefore, the part cost ( part C ) might be characterised two conditions: single part batch and optimised batch. In spite of both conditions being defined by Eq. ( 28 ), as shown in Eq. ( 29). In addition, we have also considered that the waste material is 10% of part material because of the support material, errors, purges routine, among others. waste material part material part part direct C k V C + ? ? = _ _ ? (29) In this case, it is important to note that the part volume consider a solid strategy, being ignored either weave infill, pattern infill, low density or air gap strategies (GIBSON et al., 2002;GIBSON et al., 2010). With reference to production overhead costs (Eq. ( 30)), we defined manufacturing batch time ( batch It is important to be highlighted that although additive manufacturing processes do not result in amortised tooling cost, the operational cost is amortised by the number of parts which is produced in each batch. As consequence, the maximum amortisation is restricted by the building area of machine. ? ? ? ? ? ? ? ? + ? ? ? ? ? ? ? ? ? + = s W W round s L L round N part area building down part area building down batch _ _ max_ (31) It is also important to note that this equation ignores the possibility of building several parts along z axis, resulting in a bidirectional part building matrix. For the determination of total batch time, we defined that the main components that contribute for the ) and the tool diameter ( d ), as shown in Eq. (33). In this case, the tool diameter might represent either nozzle diameter, bead width or laser beam diameter, while the part infill strategy was considered solid. Additionally, we can also define the average lead time ( part t ) per part as: ). In this case, the total amount of parts is defined by the annual demand of parts ( annual N ) multiplied by the payback period, as it is possible to be seen in Eq. ( 36). payback annual equipment machine T N C C ? = (36) With respect to the equipment cost, we identified the approximated cost of the main professional additive manufacturing machines which presented building area superior to 300x300x300mm. In this case, we considered the FDM, SLA and SLS as the main technologies to be analysed in this scenario. # d) Low Cost Additive Manufacturing in Network On the other hand, the last scenario which we analysed in this work concerns the use of low cost additive manufacturing technologies as an effective way of production. In contrast with the previous scenario, this proposal is marked by the use of low cost machines, which are also known as 3d printer. In general way, the main difference between the cost estimation model of this proposal and the additive manufacturing scenario is related to the possibility of simultaneous batches in addition to machine cost reduction. As consequence, it was possible to find that the production overhead and machine costs were the most affected component in the model. Adjusting the previous model to the number of machines ( machines N ), it can be seen that the machine cost is: payback annual machines machine individual machine T N N C C ? ? = _ (37) For this scenario, we identified the approximated cost of the main low cost additive manufacturing machines which presented building area up to 150x150x150mm. In this case, we considered only the FDM, SLA technologies. And the labour cost rate might be amortised by the number of machines which one worker can manage In addition, it was also possible to find the variation of batch lead time and the average part lead time for more than one batch as a function of the number of machines which is used in network ( machines N ), as presented in Eq. (39) and Eq.: (40). On the other hand, the lead time for the first batch is equal to the batch time. # Results and Discussions a) Injection Molding With respect to the results of this study, it was possible to characterise the cost of part as a function of annual part demand in addition to production strategy. In order to be possible to compare all the 4 studied scenarios, we established 3 main sizes of parts to be analysed: 8x8x15mm; 30x30x15mm and 60x60x15mm. For the cost estimation of molded parts, we considered that the number of cavities should be equal to 4, which is an indicated value for low scale production. The main parameters which were used in this analysis can be seen in Table 3, where the total cost of material, machine and tooling is also presented for each one of the 3 part sizes. # Table 3: Parameters for cost estimation of injection molding parts It is also important to note that the number of parts per batch directly interfere in the production overhead cost, as it is possible to see in Figure 3. In this figure, we can indicate the saturation of cost for batch sizes which are superior to 100 parts. For that reason, we selected the batch size equal to 100 parts to perform the cost analysis. As result of this cost analysis, it was possible to identify the variation of injection molding part cost as a function of annual part demand, as shown in Figure 4. In this figure, we can also see the cost of the 3 part sizes in addition to the needed capital investment for each one. In this case, besides the investment has varied from $16,000.00 to $65,000.00, the amortisation cost happens in exponential proportion. Therefore, if the $16.00 was considered a suitable part cost, the annual demand that justified the injection molding production would be 1000, 3000, 5000 parts for each one of part sizes. # Figure 4: Injection molding part cost versus annual part demand As the injection cycle defines the main lead time of parts, we can see the variation of demand and lead time for the parts, as shown in Figure 6. In this figure, it is possible to see that all the 3 analysed parts resulted in the lead time lower than the demand time. As consequence, it indicates that the machine tend to present idleness for annual part demand inferior to 200000 parts/year. In addition, if only one idle injection molding machine were considered, the minimal inventory for 200000parts/year demand and no safety stock would be 26. # b) Additive Manufacturing Services In the second analysed scenario, we investigated the feasibility of additive manufacturing services as production way. In this study, we identified the tendency of cost which is related to parts as a function of number of parts per order or batch, as it is possible to be seen in Figure 6. In this figure, the diagrams of part cost as a function of number of parts per order or batch were presented in addition to a general diagram which compile the maximum, minimum and mean values of all the three analysed technologies. It is possible to see that the cost tends to be saturated in 50 parts batch sizes, while the size of parts proportionally increases the part cost. For the FDM technology, values varied between $75.00 and $25.00 for 50 parts batch size, while the value remained near to $200.00 for a single part batch size. On the other hand, SLA part cost varied from $125.00 to $275.00 for one part batch size and from $17.00 to $50.00 for 50 parts batch size. It is also possible to establish as a general rule that SLA parts with main dimensions from 8 to 60 and 15mm of height mm tend to cost $30.00. For the SLS technology it was observed similar cost behaviour, where the cost varied according to the part size from $75.00 to $200.00 for a single part batch size. While for 50 parts batch size, the variation of cost was found between $11.00 and $75.00. Another important point that was also seen in this study is related to stock analysis. It was observed that the delivery time for additive manufacturing services is around 7 working days. Therefore, the delivery time per part is approximately 68 hours. In addition, Figure 7 presents the correlation of demand and delivery part time as a function of annual part demand. In this figure, it is possible to see that use of additive manufacturing services is feasible to be applied for annual part demand inferior to 1715 in terms of lead time. # c) Additive Manufacturing Production For the Additive manufacturing production scenario, we estimate the part cost for 3 part sizes and 3 technologies, as it is possible to be seen in Table 4. It is important to note that the machine cost is a quotation average and reflect the magnitude cost of each technology. In this table it is also exposed the maximum number of parts that might be produced by batch according to a building area equal to 300x300x300mm. Another point that is also important to note is related to the batch size, as presented in Figure 8. It is possible to see the tendency of cost saturation for batch sizes superior to 5 parts. In this figure, the variation of overhead cost as a function of part size is also presented, where the variation of cost is found between $4.00 and $45.00. It is also presented that for this same annual part demand, the part cost of 8x8x15mm parts tend to $4.00 and 60x60x15mm parts tend to $48.00. With respect to the timing analysis, we have also estimated the maximum manufacturing time per batch according to the part size. In contrast, Figure 10 presents the comparison between the lead time and demand time as a function of annual part demand. In this figure, it might be indicated the manufacturing feasibility with accordance with annual part demand. In this way, the maximum annual part demand that can be provided by additive manufacturing production in the studied conditions might be 1255 parts with 60x60x15mm, 2668 parts with 30x30x15mm and 15000 parts with 8x8x15mm. With respect to stock analysis, the minimal stock with no safety stock should be equal to 16 parts for 60x60x15mm part sizes, 81 parts for 30x30x15mm part size and 900 parts for 8x8x15mm part size, as it is possible to be seen in Figure 11. In this analysis, the maximum annual part demand for each part size was considered according to the presented before. Now for the last scenario, we analysed the feasibility of production which used low cost additive manufacturing machine in a network arrangement. In this case, the main parameters which were used can be seen in Table 5, where the machine cost and the building area dimensions are the main difference from the previous scenario. It is important to note that in this table, the production overhead cost considers only one machine in the estimation. Otherwise, the production overhead cost tends to decrease in accordance with the number of machines in the network arrangement. It is also possible to see that in comparison with professional additive manufacturing equipment, low cost equipment implied on an extremely high production overhead cost for large parts. It probably occurs because of the low raster speed and the long manufacturing time. Otherwise, the increase of machines into the network arrangement is not indicated for very small demand. At this way, fabrication of very small parts (8x8x15m) with a single machine tends to be more interesting for annual demand which is found below 1000 parts, while the small parts (60x60x15mm) seems to be more suitable to be fabricated in a network when an annual demand is higher than 200 parts/year. In Figure 13, it is presented the correlation between demand time as a function of annual part demand and the lead time which is provided by 1, 4, 8 and 16 machines in the network arrangement in addition to exposing the effect of part size for the lead and demand time. With these diagrams, it is possible to identify the production way feasibility range, where the intersection between demand and lead time marks the maximum annual demand that the production network can support. On the other hand, the analysis of part lead time has shown to be strongly influenced by the number of machines in the arrangement. In this case, the maximum annual demand of 60x60x15mm part size that might be attended by the production way varied from 450 to 6000 parts/year if the number of machines in the arrangement would be increased from 1 to 16 machines. In addition, for 8x8x15mm and 30x30x15mm part size, this number would respectively be raise to 80000 and 15000parts/year. With respect to the minimal stock considering no safety stock, this scenario implied on an inventory size equal to 4 parts for 60x60x15mm part size, 16 for 30x30x15mm part size and 255 for 8x8x15mm part size. # e) Production Strategy Comparison Comparing the results of the analysed scenarios, it was possible to identify the main differences among the scenarios in term of cost. In this analysis, it was also possible to see which production way is more suitable for each annual part demand. In order to compare the four scenarios, we identified the part cost of each process as a function of annual demand and part size, as represented in Figure 14. In this figure, it is possible to see that the most indicated production way for 30x30x15mm parts size and annual demand inferior to 1000 parts/year might be additive manufacturing services. On the other hand, for annual demand between 1000 and 3000 parts/year, the recommended production way should be additive manufacturing in an 8 machine network arrangement. In this case, it was also evidenced that injection molding was the most indicated for annual demand superior to 3000 parts/year. For 8x8x15mm part size, it was indicated that additive manufacturing services is the most indicated production way until 2000 parts/year, in addition to the network arrangement was seen to be equivalent to injection molding part cost. In contrast with this, it was found that additive manufacturing services was the most indicated for 60x60x15mm part size with annual demand inferior to 500parts/year. While the low cost additive manufacturing in 8 machine network arrangement was evidenced to be the most suitable for annual demand between 500 and 3000parts/year. For superior values of annual demand, the most indicated process was proved to be injection molding production. In addition, the part demand time as a function of annual demand was also compared in Figure 15, where all the production scenarios were shown to be feasible in term of lead time for 8x8x15mm part size and annual demand inferior to 5000part/year. Otherwise, the lead time of 30x30x15mm part size of additive manufacturing services was indicated to attend to 1500parts/year, while additive manufacturing production was to 2500parts/year. For this part size, both injection molding and low cost additive manufacturing in 8 machines network arrangement were found to support to annual demand superior to 5000parts/year. the end, as the lead time of 60x60x15mm part size tend to be longer than smaller parts, the additive manufacturing production was found to attend to 1250parts/year, while additive manufacturing services was to 1500parts/year. Additionally, low cost additive manufacturing with 8 machines in a network arrangement was identified to support to 3000 parts/year. For this part size, the only process that was found to attend to the demand time for annual demand superior to 3000parts/year was the injection molding. With respect to the minimal inventory, we can see in Figure 16 that injection molding result in the smallest inventory for small parts, while the additive manufacturing with 8 machines in a network arrangement does for medium size parts. Moreover, although injection molds and additive manufacturing services were found to imply in a constant inventory size for different part sizes the average of inventory size was marked to remain below 50 parts. In other words, no significant benefits in using additive manufacturing were seen for low scale production in comparison with conventional processes. # Conclusions In this work, it was possible to see the main differences among injection molding, additive manufacturing services, additive manufacturing with large professional machines and additive manufacturing with low cost machines in a network arrangement. In addition, estimated cost numerical models of each one of the analysed processes were developed and identified the main components that contribute for the part cost. It was possible to evidence the feasibility range of each one of the analysed processes as a function of annual demand besides being indicated the most suitable processes in accordance with the demand range. For very small demand, the most indicated production way which was found is additive manufacturing services even though it results in a high part cost. In contrast, low cost AM machines in a network arrangement were shown to be the most recommended for annual demand between 500 and 3000parts/year. It might indicate that this range of demand which was poorly covered may be attended by this proposed production way so that new business can also be created as consequence of this. With respect to the lead time analyses, it was evidenced that injection molding attend all the analysed part sizes for annual demand superior to 5000parts/year, while the additive manufacturing services was up to 1500parts/year. On the other hand, additive manufacturing with 8 machines in a network arrangement was found to attend to annual demand superior to 5000parts/year for 8x8x15mm and 30x30x15mm part size and annual demand is limited to 3000parts/year for 60x60x15mm part size. ![(REES, 1996; CHATAIN e DOBRACZYNSKI, 1997; ASIEDU e GU, 1998; FAGADE e KAZMER, 2000; ROSATO e ROSATO, 2000; NIAZI et al., 2006; RUFFO et al., 2006). Nevertheless, as the main goal of this work is to compare different production scenarios, we excluded the administrative overhead from the analysis.](image-2.png "") 1![Figure 1: Representation of the main part cost components For this study, the main cost components that have influence in the part cost were analysed, as such direct cost ( t direct C](image-3.png "Figure 1 :") ![general way, the stock flow might be analysed by the variation of delivery parts (](image-4.png "") ![Eq.(7) (REES, 1996; CHATAIN e DOBRACZYNSKI, 1997; FAGADE e KAZMER, 2000).](image-5.png "") ![estimation of machining time is characterised by the machining time of cavity volume (](image-6.png "") ![pass h is the removal depth of rough machining Feasibility Study of Digital Manufacturing Systems Applied for Medium Scale Production For the estimation of cavity machining cost, it is necessary to identify the machining labour rate ( of volumetric removal of material per time volume cavitiy t _ is the necessary time to machine the volume of cavity On the other hand, the estimation of surface cavity machining time can also be characterised by the surface area of cavity (](image-7.png "") ![factor ( d f ) and annual parts amount (N), as shown in (17) (REES, 1996; CHATAIN e DOBRACZYNSKI, 1997; FAGADE e KAZMER, 2000).](image-8.png "") ![Production Overhead CostWith reference to the estimation of basic production overhead costs (Eq. (30)), we definedmanufacturing batch time ( batch t amount of parts per batch ( batch N ) as the main cost components (REES, 1996; CHATAIN e DOBRACZYNSKI, 1997; FAGADE e KAZMER, 2000; RUFFO et al., 2006)..](image-9.png ")") 2![Figure 2: Machine shot capacity versus machine clamp force (ROSATO e ROSATO, 2000)](image-10.png "Figure 2 :") ![Study of Digital Manufacturing Systems Applied for Medium Scale Production Computer Science and Technology Volume XX Issue V Version I Year 2 020 and the shipping time ( shipping t ), as presented in Eq.: (27). Therefore, the minimal inventory might be estimate in accordance with the maximum value of Eq. (3).](image-11.png "Feasibility") ![), the overhead cost is amortised by the number of parts per batch, resulting in part cost differences (ASIEDU e GU, 1998; HOPKINSON e DICKENS, 2001; HOPKINSON e DICKNES, 2003;RUFFO et al., 2006). this study, the estimation of direct cost was mainly determined by the cost of material, whose main components are the raw material cost coefficient ( waste material ( waste C](image-12.png "") ![amount of parts per batch ( batch N ) as the main cost components.](image-13.png "") ![this case, the determination of the maximum number of parts which are possible to be produced in each batch was established by the maximum dimensions of part ( minimal distance between parts ( s ), resulting in:](image-14.png "") ![presented in Eq. (35).](image-15.png "") ![the end, the last component of part cost is related to the machine and equipment acquisition ( machine C ). As this cost is a capital investment, the total cost of equipment ( equipment C ) is amortised by the amount of parts which is fabricated during a payback period of time ( payback T](image-16.png "") ![Eq. (38), it is possible to see the formulation of this cost rate as a function of number of machines used in network (](image-17.png "") 3![Figure 3: Effect of batch size for the production overhead cost](image-18.png "Figure 3 :") 5![Figure 5: Delivery and demand time as a function of annual part demand](image-19.png "Figure 5 :") 6![Figure 6: Diagrams of part cost estimation as a function of the number of parts per order for SLA, FDM and SLS technologies](image-20.png "Figure 6 :") 7![Figure 7: Delivery and demand time as a function of annual part demand With respect to stock analysis, the minimal stock with no safety stock should be equal to 50 parts, if an annual part demand of 1715 and a batch size equal to 50 were considered.](image-21.png "Figure 7 :") cavitiymaterialcavitymaterialcavityvolumecavitiyvolumematerialvolumepassroughpassspeed 1Wall thickness (in)Wall thickness factor (t/in²)0.020-0.0626-50.062-0.1255-40.125-0.2504-3 4 © 2020 Global Journals Feasibility Study of Digital Manufacturing Systems Applied for Medium Scale Production © 2020 Global Journals ## Acknowledgments The authors would like to thank the CNPQ for financial support, as well as the Department of Mechanical Engineering of the University of São Paulo,(campus São Carlos), for providing access to infrastructure and Laboratories. * Product life cycle cost analysis: state of the art review. International journal of production research YAsiedu PGu 1998 * Product Design for Manufacture and Assembly GBoothroyd PDewhurst 1988 * The complete part design handbook: for injection molding of thermoplastics EACampo 2006 Hanser Gardner Publications 840 * MChatain ADobraczynski Injection Thermoplastiques: Moules: Ed. Techniques Ingénieur 1997 68 * DESIGN OF AN FDM POSITIONING SYSTEM AND APPLICATION OF AN ERROR-COST MULTIOBJECTIVE OPTIMIZATION APPROACH MW MCunico JDCarvalho 2013a Rapid Prototyping Journal, v.19, n.5. * MW MCunico JDCarvalho Optimization Positioning Of Machine Using Aproach Rapid Prototyping Journal 2013b v.19, n.3. * Early cost estimation for injection molded parts AFagade DOKazmer J. of Injection Molding Technology 2000 * Assisting mold quotation through retrieval of similar data MJFonseca EHenriques Digital Enterprise Technology Springer 2007 Assisting mold quotation through retrieval of similar data * Rapid prototyping for architectural models IGibson TKvan Rapid Prototyping Journal 2 2002 * Additive Manufacturing Technologies: Rapid Prototyping to Direct Digital Manufacturing IGibson DWRosen 2010 Springer 473 * Rapid prototyping for direct manufacture NHopkinson PDickens Rapid Prototyping Journal 4 2001 * Analysis of rapid manufacturing-using layer manufacturing processes for production NHopkinson PDicknes Proceedings of the Institution of Mechanical Engineers the Institution of Mechanical Engineers 2003 * Injection Mold Design Engineering DOKazmer 2012 Carl Hanser Verlag GmbH & Company KG 444 * Product cost estimation: Technique classification and methodology review ANiazi JSDai 2006 128 Journal of manufacturing science and engineering * Understanding Product Design for Injection Molding HRees 1996 Hanser Gardner Publications 116 * Injection Molding Handbook DVRosato MGRosato 2000 Kluwer Academic Publishers 1457 * Cost estimation for rapid manufacturing-laser sintering production for low to medium volumes MRuffo CTuck Proceedings of the Institution of Mechanical Engineers the Institution of Mechanical Engineers 2006 220 * Research on injection mold intelligent cost estimation system and key technologies HWang X.-YRuan The International Journal of Advanced Manufacturing Technology 2003