EFFECTS OF COMPLEXITY ON TOOLING COST AND TIME-TO-MARKET OF PLASTIC INJECTION MOLDED PARTS

Transcription

1 EFFECTS OF COMPLEXITY ON TOOLING COST AND TIME-TO-MARKET OF PLASTIC INJECTION MOLDED PARTS Adekunle Fagade, University of Massachusetts Amherst David Kazmer, University of Massachusetts Amherst Abstract The injection molding process is increasingly being used in the manufacture of complex net shaped parts. Designers are taking advantage of improvements in the process and the development of engineering materials with superior properties by consolidating multiple parts and functions into single complex parts. However, the effects of complexity on tooling and manufacturing costs as well as time-to-market of injection molded parts are still largely undetermined. This paper proposes the use of the number of dimensions that are used in detailing a part as a measure of its complexity. The metric was tested with empirical data and found to correlate well with mold cost and to a lesser extent with tooling lead-time. Introduction Plastic parts are increasingly being used in the manufacture of complex shapes mainly due to the following factors: The development of CAE simulation software able to accurately predict the shrinkage and flow characteristics of polymer in the molding process, at the product design stage. The development of engineering polymers with performance characteristics comparable to metals, and often at a lower cost. Improvements in the process control technology of the injection molding process. These factors among others have encouraged designers of discrete mechanical systems to often consider the integration of multiple parts and fasteners into fewer numbers of more complex plastic parts that are injectionmolded. The most frequently used set of guidelines for parts consolidation are the Design for Manufacture and Assembly (DFMA) guidelines developed by Boothroyd and Dewhurst [1]. One significant benefit of DFMA is the considerable savings in assembly cost from fewer parts that need to be assembled. Other inherent benefits of DFMA are the encouragement of teamwork between design and manufacturing, and the improved product reliability from the reduced probability of system failure because of component failure. The application of DFM guidelines is said to lengthen the concept development time but helps to shorten the other stages of product development. However, during the parallel development of a product s components, the earliest time-to-market, is determined by the most complex part. The product development stages of a plastic part are assumed to consist of the overlapping designing, prototyping, tooling, and manufacturing stages. The effects of part complexity on the tooling stage are investigated in this research. Time-to-market determines to a large extent the profit realizable from a high-tech product over its lifetime. According to a McKinsey and Company study, a high tech product that reaches the market six months late, even on budget, will earn 33% less profit over five years. On the other hand, finishing on time but 50% over budget will reduce a company s profit by only 4% [2]. Another motivation of this research is that moldmakers have been unable to predict the effects of complexity on mold manufacturing cost. The extreme competition faced by mold-makers is such that the costs of increased job complexity and shorter lead times often cannot be included in the original bid price, or the job will be awarded elsewhere. We are forced to spend an enormous amount of time reviewing these jobs even after the mold designs are finished, to see if we can trim costs. Still, any changes affecting cost or delivery must not affect the quality of the mold we build, said an executive of a mold making company [3]. In this work, multiple regression models for predicting the effects of part and system complexity on tooling costs as well as tooling lead-times at the design stage are empirically developed from field data. The empirical study used the quotes submitted to a custom injection molder by local and overseas moldmakers for thirty plastic parts with varying sizes and complexity. A high correlation was found between the mean of the quotes received for each part and a function of the part s basic envelope volume and the number of dimensions required to fully detail the part. The linear regression correlation also improved significantly when the effects of three other parts attributes that contribute to injection mold tooling costs were included. Previous Work on Tooling Cost Estimation Two methods published in design literature that address the problem of estimating mold tooling cost at the design stage are the Dixon and Poli [4], and the Boothroyd and Dewhurst [1] methods. The two methods agree that part s geometrical complexity is a significant contributor to its tooling cost but evaluate complexity differently.

2 The Dixon and Poli method estimates the relative tooling, material, and processing costs of an injection molded part from look-up tables. These costs are estimated relative to the cost of tooling for a simple reference part. The reference part used is a flat disc with outside diameter of 72 mm and inside diameter of 60 mm. The approximate tooling cost for this reference part, based on costs, is $7000. This includes about $1,000 in die material costs. Seven attributes that can be determined from the part at the design configuration stage are used in evaluating a part s basic complexity, C b, from a look up table. These attributes classify the part by its size, shape, number of walls with undercuts etc. Two multipliers of C b, are also evaluated from look-up tables. They are the subsidiary complexity factor C s and the tooling and tolerance factor C t. C s is a function of the number of form features in the part s cavity and number of undercuts. C dc = C b C s C t (1) C d = 0.8 C dc C MB (2) Where C dc is the relative mold construction cost, C d is the relative mold cost and C MB is the mold base cost. A current mold cost for the reference part is required to use the model. In the determination of subsidiary complexity, the classification of some cavity detail features as regular or irregular, and the qualitative classification of undercut complexity as extensive or not extensive requires the judgement call of the estimator. The Boothroyd and Dewhurst (B-D) method uses empirically derived formulas and estimated manufacturing points to estimate the times for the different tasks that are carried out in transforming a purchased mold base to a finished mold. The sum of these times are then multiplied by an average shop rate R, to get an estimate of the tool construction cost. An empirically derived formula is used to estimate mold base cost as a function of the area of mold base cavity plate and the combined thickness of the cavity and core plates. Mold tooling cost is then the sum of moldbase cost and mold construction cost. The authors refer to a rule of thumb in mold construction, which is that the purchase price of the mold base should be doubled to account for the custom work on it [5]. CTool C MB R ti (3) where: C MB = Estimated mold base cost. R = Average mold shop hourly rate ($). t i = Estimated times for the different mold construction operations including additional times due to complexity, surface finish, tight tolerance and parting surface complexity. The B-D method calculates part complexity as a sum of inner and outer surfaces complexities. The surface n i complexities are estimated with an empirically derived formula that uses number of holes, depressions and surface patches. The generalization of all possible design features into the three categories limits the sensitivity of the B-D complexity index. Also the enumeration of surface patches is difficult and uncertain for moderately complex part, which may have blending surfaces and many protruding rib features. Automating Evaluation of Part Complexity In view of the difficulties with the two previous methods, a research was conducted for tooling cost drivers and complexity metrices that can be evaluated from a CAD product data at the design stage. Preference would be given to quantitative metrices that would not require the subjective judgement of the designer. Complex systems are known to consist of finite variety of interacting elements. According to Scurcini [6] the number, variety, types, and the organization of elementary components drive the complexity of a technological system. Since form and shape features constitute the basic components of a plastic part, an enumeration of the features in a designed part could be functionally related to its complexity. Part complexity together with other mold cost drivers can then be used to determine its tooling cost etc. The problems with the use of features in cost estimating are the same as some of the problems faced with feature-based design. Three approaches are frequently used for product design with feature: interactive feature recognition, feature extraction, and feature-based design. Of the three, only feature-based design is easily amenable to easy and automatic enumeration of all the features in a part from its CAD data. Feature extraction requires complex algorithms that have only been successfully implemented in the recognition of simple feature profiles. Injection molding form features were classified to conform to the Form Feature Information Model (FFIM) [7] established for the Standard for the Exchange of Product Data (STEP), Figure 1. However, since the designer has the freedom to define application type features, as long as they fall within the FFIM classification, cost estimation models built on a fixed number of features are soon rendered obsolete. In our research, the initial approach was to assign every type of feature a price tag based on the cost or difficulty of reproducing the feature in an injection molded part. A process to count parts features from blueprints was initiated. However, correctly identifying and classifying all the geometrical features of a part from its blueprints or even from a physical sample is not a trivial task. Two engineers would prepare different features list given the same part. Problems such as whether to classify a set of parallel protrusion features as ribs or grooves existed.

3 EXPLICIT FEATURES List of two-dimensional shape elements FORM FEATURES DEPRESSION DEFORMATION PROTRUSION AREA FEATURE * Not Applicable * Boss * Groove * Disk * Thread * Hinge * Flange * Marking * Pocket * Gusset * Gear teeth * Step * Projection * Surface Finish * Ribs * Texturing * Snapfit * Tab IMPLICIT FEATURES PASSAGE * Hole * Slot * Window TRANSITION FEATURE * Edge blend * Corner blend Figure 1: A Classification of Injection Molding Features However, for the purpose of cost estimating, we only need to know the complexity created by the number and variety of features. We surmised that the number of dimensions that are used to define a feature is a measure of its complexity since the more the dimensions that are used to define a feature the more difficult it is to manufacture the feature. Every dimension represents an additional point to check or a setup to make in the manufacturing of the mold. This reasoning is then logically extended to the total number of dimensions required to completely define the parts model. This is particularly true in constrained-based type modelers, which include most 3-D modelers, where for example a solid block would require three dimensions, its height, base, and width to uniquely define it. The block is then said to have a complexity of three. Collection of Field Data A custom injection molder, in Western Massachusetts, assisted in this research. Original equipment manufacturers (OEM) submit requests for quotes (RFQ) to this company. The company in turn sends out requests for tooling quotes to moldmakers locally and overseas. Records of quotes received and sent out are kept in filing cabinets. Seventy-five mold tooling quotes of single cavity molds for thirty of the parts that the company has quoted for in the past three years were selected for analysis from its records. According to the tooling engineer at the custom molder, past performance is a critical factor they considered in giving out tooling contracts to moldmakers. The size of the part, its complexity, number of slides, gate type, surface finish, injection and ejection systems, are some of the factors that are considered in the rough estimation of a part s tooling. When the quote from a toolmaker falls outside a range estimated by the tooling engineer, it could be due to one of three reasons. If the mold quote is too low, it could be that the toolmaker failed to consider the need for slides or other factors. These factors may not have been too apparent from the blueprint or CAD model. In this case, the molder s tool engineer tries to confirm that the toolmaker considered all design specifications. If the quote is too high, it could be that the moldmaker is at its capacity and just wants to submit a quote anyway. It could also be that the tooling engineer overlooked some design specifications. However almost in all cases, there is always the postdesign stage cross communication among the three parties comprising of the design engineers of the product developer, the tooling engineer of the molder, and the toolmakers. Engineering changes, usually minor, that may reduce tooling cost and/or facilitate molding, are suggested to the product designers and are either accepted or rejected. The thirty parts, their mean mold quotes (MMQ), mean estimated tooling lead times (MLT), and their geometrical attributes are as shown in Table 1. Only RFQs accompanied by blueprints that have adequate detailing for tooling were selected. A few quotes that were much higher or much lower than the average quotes for the same part were not selected due to the probability of over or under estimation, as mentioned previously. In Table 1, the envelope volume in cubic centimeters measures the size of the part. This is the volume of a rectangular box that completely encloses the part. Even where the projection is isolated, the envelope volume still determines the size of the mold base and to some extent the manufacturing work required to make the mold. Table 1: Quotes and Attributes of Observed Parts # MMQ MLT Size (cc) # Dim. # HF HT ($K) (wk) Acts Y N N N N N Y Y N N N N N Y Y N N Y Y N Y Y N N Y N Y Y Y Y Y Y Y Y N N N N N N N Y N N N N N N N N N N N Y Y N Y Y Y Y The total number of dimensions (# Dims.), is the number of dimensions required to unambiguously define the part. In the current work, these were enumerated by counting all the dimensions on all blueprints that accompanied the RFQ. All dimensions in all views; elevations, sectional, and detail, were counted. The total number of dimensions is seen to be a good measure of a part s complexity.

4 When a view represents a repeated feature the number of dimensions is multiplied by the number of times the feature is repeated. Usually, such views if labelled in accordance with ANSI Y14.5M dimensioning and tolerancing standard [8], show how many times the feature is repeated by a number and an X as in Figure 2. Figure 2: Detailed sketch of a low complexity part Table 2: Counting Dimensions of Gasket Disk Envelope Size: 160x160x5 mm 3 = 128 cc Type and number of dimensions Circular hole features 1x8 = 8 Angular spacing of holes = 8 Dimensions of slots 2x4 = 8 Diameter of center hole = 1 Radial distance of holes = 1 Ref. angle from center line = 1 Envelope dimensions = 2 Chamfer radii (transistions) 4x4 = 16 Total number of dimensions = 45 Edge and corner chamfer and fillet radii, that are classified as transistion features in the form feature information model of STEP, were then counted separately. The number of these transition dimensions were subtracted from the total number of dimensions to obtain the number of early dimensions. Early dimensions represents the number of dimensions available at the early stage of design before the design is fully detailed. As the design progresses and all chamfer and fillet radii are specified then the total number of dimensions are known. The number of actuators (# Act.) is the total number of separate mechanisms that have to be constructed into the mold to permit molding of internal and external undercuts, and screw features on the part. Undercut features that lie on the same wall of the part and that are within 75mm distance of each other are are assumed to require one slide mechanism. Every screw feature is assumed to each require a separate unscrewing mechanism. The parts with Y (Yes) under the columns labelled HF and HT in Table 1, are the parts with high polish finishes and tight plastic tolerances respectively. Parts with surface finish specifications of SPI A1, A2, and A3 or that are textured on more than 25% of their entire surface areas are classified as having high polish finishes. Parts with surface finish of SPI B1 or less on more than 75% of their surface areas, are classified as having normal finishes. Plastic tolerances are specified as percentages of overall lengths. Due to shrinkage characteristics of polymers, longer parts are normally specified with larger tolerances. Molds are manufactured within a small percent of the tolerance allowed on the plastic part. A cut-off value of 0.07% of absolute percentage tolerance per unit length was used to classify the observed parts as having tight or normal plastic tolerances. Parts with absolute percentage tolerance per unit length less or equal to 0.07% were classified as having tight tolerances, while those with greater values have normal tolerances. The decision was guided by a table of dimensional tolerances allowed to mold makers [9]. A general tolerance level was determined for each part by observing the tolerance specified for more than half of the part s dimensions. For example, if more than half of the dimensions are specified with tolerances of 0.05mm then the general tolerance level is 0.05mm. Results and Discussions Multiple regression analyses were performed with the mean mold quotes and mean lead-times as dependent variables. Combinations of the other parts attributes from Table 1 were used as the independent variables. Other parts attributes measurable from a blueprint or CAD model were examined, such as part projected area, material volume of part, number of critical-to-function dimensions, variety of dimensioning and tolerancing annontation. Only the best correlations are reported in this paper. In the summary outputs of the regressions, the sample coefficient of multiple determination R 2, is the proportion of the total variation in the values of the dependent variable that is explained by the independent variables. R 2 Explained Variation Total Variation R 2 can take on values between 0 and 1. The closer R 2 is to 0 the worse is the fit of the regression plane to the data; the closer it is to 1, the better is the fit. The regressions were done at 95% confidence level. The abbreviated summary outputs follow: MMQ = (Size) ( # Dimensions) R 2 = (5) Equation 5, shows that size and number of dimensions explain 87% of the variation in mold cost of the sampled parts. The intercept, 28,300, represents on the average the lower bound on the mold costs. Three other parts attributes; number of actuators, high surface finish (HF) and high tolerance (HT), are then (4)

5 included in the regression analysis, with the latter two having only 0, 1 dummy variables. The model now explains 91.1% of the variations in the mold costs, Equation 6. MMQ = (Size) + 30(# Dimensions) (# Actuators) (HF) (HT) R 2 = (6) The mean tooling lead-time has a lower but still very significant R 2 value when regressed against size and total number of dimensions (complexity), as shown in equation 7. The imperfect correlation may be due to other molder specific factors, such as availablity of excess capacity, or willingness to expedite a job to gain a customer. The minimum of 13 weeks can be considered the minimum lead time that molders would normally take to tool a simple part. Tooling Lead-time (weeks) = (Size) (Total number of dimensions) R 2 = 0.7 (7) These results show the much higher effects of increases in complexity, as measured by the number of dimensions, on tooling cost and tooling lead-time compared to the effects of size increases. Equation 5, shows that every 100-count increase in number of dimensions, which is a normal phenomenon when parts are consolidated into complex parts, increases tooling cost by $4560, and tooling lead-time by 5 days. A comparable increase in mold cost due to size increase is only possible if the size of the part is increased by 5,600 cc, a six-fold increase if starting with a 1000 cc part. Equation 7 also confirms that size has very negligible effect on tooling lead-time compared to part complexity, a ratio effect of 1:127. The results show that consolidation of parts is preferable when the parts to be combined have low complexity. Consolidating two already complex parts into a more complex piece may increase tooling cost and tooling lead-time drastically. The cost incurred in higher tooling cost, and lost sales due to late market introduction may surpass the benefits expected from the parts consolidation. When timely market introduction of a product is critical to its life cycle profit, it is preferable to develop and parallel-tool simple components for automatic or manual assembly than to combine components into a complex piece. The single complex tool may take longer to tool and may cost more than the individual tools put together. Consolidation may later be done when demand exists and new sets of tools are being ordered for large production runs. This current research has only looked at effects of part complexity on tooling cost. Further work is being done to also investigate the effects of complexity on processing, assembly, usage and other stages of a product life cycle. The models described are easily developed by any organization that has historical data on mold costs. It is easily updated by re-evaluating the regression coefficients when adequate current data is available. The model is currently being implemented on a SolidWorks 98 modeler. Conclusion A new way for evaluating a part s complexity at the early stages of its life cycle using the number of dimensions from its blueprints or CAD data has been proposed. The results from empirical data show that this metrics correlates well with mold tooling cost and the lead-time to develop the tool. The analyses also identify part complexity and size as two significant drivers of tooling cost and lead-time. New rules are being created to augment current design for manufacturing guidelines. References [1] G. Boothroyd, P. Dewhurst, and W. Knight, Product design for manufacture and assembly. New York, NY: Marcel Dekker, Inc, [2] C. Charney, Time-to-market: Reducing product lead time. Dearborn, MI: Society of Manufacturing Engineers, [3] V. E. Wigotsky, Mold Making & Mold Design, in Plastic Engineering, vol. 52, 1996, pp [4] J. R. Dixon and C. Poli, Engineering design and design for manufacturing, a structured approach. Conway, MA: Field Stone Publishers, [5] A. Schuster, Injection Mold Tooling, presented at Society of Plastic Engineers Seminar, New York, [6] G. B. Scurcini, Complexity in Large Technological Systems, presented at Measures of Complexity, Rome, [7] J. J. Shah and M. T. Rogers, Functional Requirements and Conceptual Design of the Feature Based Modeling System, Computer Aided Engineering, pp. 9-15, [8] ASME, ANSI Y14.5M, American National Standard Engineering Drawings and Related Documentation Practices: Dimensioning and Tolerancing. New York: American Society of Mechanical Engineers, [9] D. V. Rosato and D. V. Rosato, Injection Molding Handbook. New York: Van Nostrand Reinhold Company Inc., Keywords Mold Cost Estimating, Complexity, Feature Based Modeling, Plastic Injection Molding