24 Manufacturing Analysis: Some Basic Questions for a Start-Up Company Chap. 2 Ftpre2.2 The market adoption curve, modified to include several products cloned sheep would be way down in the bottom left corner.) Whether or not one of these technologies will climb the market adoption curve depends on the following: • The cost benefit to the consumer • The robustness and usefulness of the technology to the consumer • Complementary uses for the consumer (in the case of consumer electronics this may well mean other applications or software that can run on the new device) • Whether or not the device falls in line with industry standards Such projects bring out the difficulties associated with launching radically new products. Geoffrey Moore (1995) introduced the phrase crossing the chasm to emphasize this (Figure 2.3). In the early stages of a product's life, there will always be some measure of a market. There is a small group of consumers who love tech- nology enough to buy a new product, no matter how useful it is.perhaps just for amusement or to be able to show off to their friends that they have the latest "cool" thing on the market. But this early market of impressionable "technology nuts" or "technophiles'' only lasts so long. The real growth of the product depends on the first bullet above, namely, the cost benefit to the average consumer. This begs the question, How can a product survive across this chasm between the early enthusi- asts and the real market? Geoffrey Moore goes on to use a "head bowling pin" analogy for capturing this phase of stable growth. As an example, he chooses the now-familiar personal pager. Apparently, medical doctors were the first real market adopters of the pager, and once the general public saw how useful they were, the product "crossed the chasm" and accelerated rapidly up the S-shaped curve in Figure 2.2. 'Autos Steel Personalcomputers Telecommunications and networking (Mobile-wireless technologies Td"-}J'''StalC'' technologies • 2.2 Question 1: Who Is the Customer? 25 Flgure2.3 The concept of "crossing the chasm" (from Geoffrey Moore, 1995). Purchasing by compulsive "technophiles" Maturity tmontha) 2.2.2 Inevitable Trade-Offs between Cost, Quality, Delivery, and Flexibility (CQDFI This brief and informal introduction illustrates a wide spectrum of market opportu- nities with a wide variety of quality levels and consumer choices.Thus, inevitably, any company must face the rising costs (C) of adding higher quality (Q), faster delivery (D), or having a more flexible (F) manufacturing and supply line (Figure 2.4). Despite the rising costs of quality, delivery, and flexibility, today's customers are more informed and they expect more than they used to. For example, they expect a Pentium chip to perform perfectly and yet be competitively priced. So how do man- ufacturers respond to this? At first glance, manufacturing seems to require too many trade-offs to simultaneously achieve high quality and fast delivery in small lot sizes, while still maintaining low cost. For example: • Supercomputers and high-end, graphics-oriented UNIX machines cannot be fabricated quickly and in small lot sizes and then sold at CompUSA in a price range suitable for families and college students. • A Lamborghini Ora Ferrari cannot be fabricated quickly and in small lot sizes and then sold at the price of a Honda Accord or a Ford Taurus. However, on closer inspection, the marketplace does set up a compromise sit- uation between the engineering constraints and the broader economic goals. This compromise involves a reliable prediction of market size and type, supported by design and fabrication systems that can appropriately address each market sector. In this way, luxury automobiles will always be more expensive than economy cars, but in the luxury car sector, the industry leaders will be those who refined their design and manufacturing skills the best to give a high quality-to-cost ratio. Consumers are, Purchasing by real market adopters I 26 Manufacturing Analysis: Some Basic Questions for a Start-Up Company Chap. 2 Figure2.4 Schematic graph of rising costs for adding more quality to the product, faster delivery (schematically measured on the.r axis by the shortness in days or weeks to deliver). and more system flexibility (CQDF).Also compare with Figure 2.10. of course, highly influenced by cost, which must be clearly related to the benefits that accrueThe relationship between cost and quality will thus be examined in the next two main sections. The relationship between cost, delivery, and flexibility will follow. 2.3 QUESTION 2: HOW MUCH WILL THE PRODUCT COST TO MANUFACTURE ICI? 2.3.1 General Overview to Calculating the Manufactured Cost of a Product Assume that N "" the number of devices sold over the life of a particular product. The cost of each product is,in the simplest terms: The cost of an individual product, C = (DIN + TIN) + (M + L + P) + 0 (2.1) In this general equation the terms are: • D = design and development costs.This includes conceptual design, detailed design, and prototyping. • T = tooling costs. This particularly includes costs for production dies. These first two costs are amortized over the number of products made and therefore divided by the number made, N. The next three costs are consumed by every product, and finally there are the overhead costs. • M = material costs per product. • L = labor costs per product for operating machines, assembly, inspection, and packaging. • P = production costs per product associated with machine utilization, including loan payments for machinery. • 0 = overhead costs. This might include rental space for offices and factory, computer networking installation and servicing, phone installations and Quality or fast delivery or Jlexibilny Q,llD,F 2.3 Question 2: How Much Will the Product Cost to Manufacture (e)? 27 Figure 2.5 Cosl breakdowns for manufacturing, similar to Equation 2.I-not to S<;:1I1e(courtesyofOatwald,1988). monthly costs, electrical and other physical services, general advertising, and general support staff This cost is more of a flat rate for all activities and must eventually be allocated per component. Reviewing this list of potential over- heads, it is not surprising that many small companies build the first prototypes in the garage or basement of one of the founders. Another view of the above equation is shown in Figure 2.5 from Ostwald (1988). It provides an overview-not to scale of all the costs involved in creating, manufacturing, and selling a product. At the bottom of the chart, the prime costs are for the direct labor and materials. Manufacturing overhead is then shown, which accounts for the labor costs to run the overall factory, consumable materials such as fixtures, and other charges for maintenance and the like. On top of this are many other overhead type charges for running the organization. Included in the engi- neering costs are the design and development costs for a particular product in Equa- tion2.1. Time-to-market is a phrase that will be used in this book to measure the time between (a) the first moment an engineer starts to bill his or her time to the company Profit Selling Contingencies Enll;ineerinJl; General and administrative Manufacturing charges Indirect materials Indirect labor Direct labor Direct materials Conversion costs Overhead =. 'PrUne cost. COSI of goods manufactured 'Cost of manufacturing, development.and sales, 28 Manufacturing Analysis: Some Basic Questions for a Start-Up Company Chap. 2 at the "conceptual product" stage inFigure 2.1and (b) the moment the product issold to the first customer and some revenue occurs.Note in Figure 2.1 that the time-to- market extends all the way around the diagram to the "9 o'clock position." Colloqui- ally speaking, there are many opportunities to run up huge debts and "go broke" before selling anything. Section 2.5 on delivery therefore examines the time-to- market issue in more detail. 2.3.2 Specific Costs for Individual Manufacturing Processes Given these introductory remarks, an interesting question now arises: Are some manufacturing processes cheaper to use than others? If so, since the designers and the production planners have a wide variety of possible methods to choose from, why not choose one that gets the cheapest and most predictable results? The book now reviews the manufacturing processes of the "mechanical world" and discusses the constraints that arise. For a reader unfamiliar with mechanical manufacturing processes, a brief taxonomy is given in Table 2.1. More details are included in Chapters 4, 7,and 8. The Manufacturing Advisory Service at <cybercut.berkeley.edu> shows illus- trations and related information. The first entry is the solid freeform (SFF) family that appeared after 19R7 with the introduction of SLA. The other families have been categorized in that way by the Unit Manufacturing Process Research Committee of the National Academy (Finnie, 1995). The analysis that follows is based on personal experiences in the metal pro- cessing industries and could probably be refined to suit other domains of manufac- turing such as semiconductors. Also,the texts by Kalpakjian (1997) and Schey (1999) cover similar material in their last fewchapters. Some of their diagrams and concepts are integrated (with acknowledgment) into this text. Other work being expanded at the time of writing is a "process selector" by Esawi and Ashby (1998). 2.3.3 Manufacturing Advisory Service (MAS) at <cybercut.berkeley.edu> Picture any of the standard mechanical components in a lawn mower, washing machine, or even a simple can opener. •Should the mechanical component be completely machined from a solid block? •Should it be near-net-shape cast or forged and then finish machined? •Should it be welded or riveted together from several pieces of standard stock? These are only three of many possible manufacturing routes. The route to choose usually depends on a rather complex interaction between guiding eubpnncl- pies of manufacturing process selection shown inTable 2.2. 2.3 Question 2: How Much Will the Product Cost to Manufacture (e)? 2. Family TABLE2.1 A Brief Taxonomy 01 Mech8nlca/ Manuf&cturing Proceue. (Courte8y of Flnn ••,l~) BriefexplanalionofprocessesProcesses LSolidrree.rorm fabrk:atioo(SFF) """" alfeJlorlzed. Iaye"" 2 •••••.••••••• p- 5. DeformaUon I. Stereolithography 2. Selective laser sintering(SLS) 3. Fused deposition modeling (FDM) 1.Drilling 2.Milling 3.Thrning 4. Grinding 5. EDM and ECM 1.Casting 2. lnjection molding of plasrics Icouki includeFDM) 1.Coalings 2.SurfaceaUoying 3. Induced residual 1.Rolling 2. Sheet drawing 3. Extrusion 4. Forging 1.Powder metals 2.Cumposites 3.Weldinglbrazing 2.3.4 Batch Size 1.Stereoltthography (SLA) uses a laser to photocure liquid polymers. 2. Selective laser sintering (SLS) uses a laser to fuse powdered metal. 3. Fused deposition modeling extrudes hot plastic through a nozzle. Like "mini- toothpaste, hot extrusion," it builds up a model These processes remove shapes from a solid block. called the stock. LA simple drill from the hardware store creates holes of different depth and diameter. 2.A milling cutter has a flat end and can cut on its sides. It can carve out flat pockets in a block to make an "ashtray." 3.Turning is done on a lathe. The stock is round. The turning tool passes up nnd down the rotating stock removing layers. "A round bar can become a sculpted chair leg." 4. Grinding/polishing use abrasives to remove thin layers of metal to greater accuracy than processes 1 4.5. Electrodischarge machining uses electric arc: Electrochemical machining uses charged chemicals to remove fine layers. I. In casting, mollen metal is poured into a hollow cavity in sand, initially created from a mold. 2.lnjeclion molding "shoots" hot liquefied plastic into a mold. 1.Hard surface coatings can be deposited chemically or physically on softer or tougher substrates-c-chrome plating is an example. 213.Alloying or shot blasting toughens .ur{a"es. 1.Slabs can be rolled down to strip as thin as everyday "aluminum kitchen foil." 2. Such sheets can be cut and ben! into office furniture, filing cabinets, or soup cans. 3. Like large-scale hot toothpaste, extrusions of different cross sections can be made if the die (the hole at the end) is a premade shape-using milling or EDM. 4. Hot or cold forging involves "slamming" metal into a die cavity. The metal stock plasticaUy deforms to the desired shape. 1.Powdered metal is formed in a die and then aintered to give full strength. 2. Layers of different carbon fiber sheets are an example of composite materials. 3.Welding involves local melting and "mini casting together" of adjoining plates. Brazing uses solid-state bonding between a filler melalannth ••twnplales When considering these subprinciples for a suggested manufacturing method, it makes sense to begin with the criteria that make the most impact on the costs. Usu- ally batch size,strength, geometry, and tolerance are the four most significant factors. 3.PbllIe-dumge - 6,COlllOlklation • - 4. strumue-dumge - 30 Manufacturing Analysis: Some Basic Questions for a Start-Up Company Chap. 2 General principles driven by designer TABLE2.2 Subprlnclpl ••of Manufacturing Process Selaetfon Implied considerations for Ihe manufacturing process(mechllJljca1manufactqriogproce.~.es) 2. Batch sbe 3. Streagth lUld welpt nlated to material dtoke 4. Geometry 5. Tolertll;lccs 6. Product 6fe 7.LeadUme 8.Deslpror usembly(DFA) Cost is driven by all the principles below. It is reviewed in the tel(t in relation toallpararneters. Only the SFF, rapid-prototyping processes, machining, and possibly casting are suitable for "just one" component. For a structurally useful product, only CNC machining and casting are realistic, but FDM can provide an alternative for low strength ABS-plaslic parts. For two to five components, CNC machining is likely. As batch size increases, snort-run, plastic injection molding or casting is considered. The cost of the die or mold is always a key factor A need for high strength drives the choice toward metal processing over plastic molding, Even higher strengths/performance will force the choice toward forging or machining over casting. A need for light weight may drive the choice to plastic, aluminum, or titanium. In general, costs will increase when high strength and performance are required. Materials such as titanium are very costly. Wide, thin cruss ~eL"tions will drive choice toward blow molding for plastic parts or sheet-metal forming for metal parts. "Chunky" cross sections drive the choice to castinglforginglmilling operations."Cylindrical" cross sections drive the choice to turning. In general, parts with complex geometries will cause high manufacturing costs. In small batches, the CNC programming and execution times will be long. 10 large batches, dies are expensive to create and operate. Tolerances tighter than +1- 50 microns (0.002 inch) will begin 10 drive the choice to the machining/grinding/polishing family of processes. Grinding and polishing are expensive operations-if possible the designer should avoid such fmishingcosts. A desired long service life will probably drive the choice to metal processing over plastics. Design constraints on fatigue properties may require special tooling and/or lapping processes. Longer desired service life will almost certainly increase costs because of more or better materials used, improved surface finish, and more careful design optimizetlona To deliver the part to the designer quickly, production planners hope to use standard processes and tooling. Weird designs might require special tools and lengthy hand assembly and finishing. Special tools and fixtures will rapidly drive up the costs and delivery time. Any process that requires a die or mold will be more expensive and will take longer than machining or SFR The way in which an individual part is mated or fixed to another one in an assembly is also important. Welding, riveting, and bolting costs are high, and these processes are often poorly controlled. Thus, cost reductions may result from new assembly or single-piece manufacturing methods. 2.3.4.1 Batch Size of 1 If only one component isneeded, then one of the rapid prototyping methods such as stereolithography (SLA), selective laser sintering (SLS), fused deposition modeling (FDM), CNC machining, or casting is the obvious manufacturing choice. The first three listed are collectively known as solid freeform fabrication (SFF). Note that SLA produces a low strength pbotopolymer model, not a structural part. SLS can LC ••• 2.3 Ouestion 2: How Much Willthe Product Cost to Manufacture (e)? 31 create low strength metal parts, and FDM can create low strength ABS plastic parts. The Sff and CNC machining processes are economic for one-off components because they do not require the time to make an expensive die or mold prior to pro- duction. Casting is also possible for large, complex single components that cannot feasibly be made by machining. However, an expensive mold of wax or wood is needed. This creates the shaped cavity in sand before the molten metal is poured in. Actually, if the batch size is as low as only one or two components and if the part geometry is simple, a skilled craftsperson might even use a manual milling machine or lathe because the programming time for a CNC machine might not be worthwhile. However, if the part geometry is complex, it is worth the time spent to program the CNC machine even for one-off components. The reason for this is rea- sonably subtle: on a manual machine, if an error is made near to the point when the part is almost complete, "all is lost" -c-not only the piece of work material but all the time that was invested up until that point. But on a programmable machine, if an error is made toward the end, there may be a scrapped workpiece but all the geo- metrical programming steps that were invested up until that point are stored in the computer. This of course begs the question, How is part complexity measured? One answer might be that part complexity increases with the number of lines of CNC code needed to machine the part. For a batch size of one, if the desired part has a complex geometry, SFF rather than machining processes will be used. An object that resembles a doughnut will be easy to make by the SFF methods but almost impossible to make by CNC machining unless it can be made in two halves and joined along its equator. Speaking very gen- erally, the more complex the shape, the more likely SFF processes will be used. How- ever.component strength is a vital consideration. CNCmachining and casting are the only viable alternatives if high strength is needed. Selective laser sintering (SLS) can produce a metal part, but it will he weaker than one produced by machining. Fused deposition modeling can produce a plastic part with reasonable strength-but, again, one that is weaker than a part from plastic injection molding. 2.3.4.2 Batch Siz~ 2 to 10 If only a few components, say 2 to 10,are needed, then CNC machining is today the most likely choice, unless the geometry is highly complex and sculptured, in which case a batch of SLS or FDM components might be realistic and cost effective. 2.3.4.3 Batch Size 10 to 500 A batch of 10to 500 might well be done by CNC machining. This batch size is begin- ning to enter the realm of a productiun shop rather than a custom prototyping shop (Figure 2.6). However, manual transfer of parts between machines will be quite likely since this batch size is not sufficient to warrant investments in automation. It should also be noted that ifthe customer isincreasing its order from just one to this higher batch size,it might be best to "backtrack" and make a good stereolithog- 'These numbers for batch size are very approximate and depend on several factors including part complexity. The ranges in these next few pages deliberately overlap. 32 Manufacturing Analysis: Some Basic Questions for a Start-Up Company Chap. 2 flJure 2.6 Trends from manual, to CNC, to reprogrammable systems (FMS). and 10"harder automation" (courtesy of Ostwald,1988). raphy pattern that willthen be used for castings.'Thisoption will work ifthe desired tol- erances are within the capability of both the SLA master and the subsequent casting process, rather than machining. 2.3.4.4 Batch Size 100 to 1U,OOO CNC machines, arranged in larger complexes called flexible manufacturing systems (FMS), willbe favored as batch size grows.Perhaps robots Ofautomated guided vehi- cles (AGVs) will be used to move parts from one machine to another. Efficiency will be very dependent on the communication software needed to orchestrate the system. However, batch sizes of several thousands will begin to warrant the Iabrica- tion of an expensive die that can rapidly punch out products by a cold forging or stamping process. While processes that require a premade die or mold are rarely, if ever, used for one-off or short batch runs, the cost of the die can be amortized over these larger funs and the COSl of the tooling die per component decreases (parameter T in Equation 2.1). Sands (1970) presents a comprehensive analysis for different forming processes showing at which batch size the use of a die becomes efficient. Die costs and manufacturing system costs increase from left to right in Figure 2.6.This cost factor places an important responsibility on the designer. In an ideal situation the newly designed component will be made on existing factory floor machinery, readily leading to an "off-the-shelf" automation solution. In the best case, existing fixtures and even some parts of existing dies will also be reused. IJoblotindust~ Moderate quantity industries Mi"ss production industries] Mechanization ~grammable , automatio~ Flexible work station csn Work station Production -systems 2.3 Question 2: How MuchWillthe Product Cost to Manufacture (Cl? 33 2.3.4.5 Batch Size 5,000 to millions As batch size increases. automation plays a bigger role. However.for extremely large batch sizes, it might even be economic to revert to noncomputerized machines. Speaking colloquially, this batch size moves into the realm of "ketchup in bottles." where fixed conveyor lines pump out the same product day in, day out. This is often referred to as fixed or hard automation, literally because "hard stops" are fixed in place with wrenches.These hard stops establish the positions where components rest in place while being filled or labeled. Some basic computer control and sensors are needed to keep things on track, but reprogramming will not be needed. 2.3.5 Material Choice The material that the designer chooses for the part willbe influenced by weight con- siderations, cost factors. and desired strength. This desired strength of the finished object isobviously a key factor. Even though metals are generally stronger than plas- tics. injection molded and thermoformed plastics are preferred for ordinary con- sumer products such as household appliances, consumer electronics, and many automotive products. In general manufacturing costsfor plastics are lower than those for comparable metal products. This is because plastic forming requires much lower forces than metal forming, and so the machinery is cheaper, the dies are less complex, and the labor costs are often lower.However, critical components such as transmission gears need to be made from closed die forging blanks to obtain a well-distributed grain structure. Finish machining completes the critical gear tooth involute profiles. This issue of basic strength is obviously related to the primary material that the object is to be made from-not onlyits basic data-book strength but also its purity,heat treat- ment, and in-process characteristics. The latter include the work-hardening proper- ties of metals and the shrinkage characteristics of plastics.This is also an important moment in the text to reemphasize that solid freeforrn (SFF')prctotyping techniques such as stereolithography create plastic components from a photo-curable liquid. This material from the SLA bath is by no means as structurally sound as standard plastics such as ABS and polystyrene. FDM can create ABS parts of reasonable strength, but not with the same structural integrity as injection molded ABS. 2.3.6 Part Geometry The product's geometry embodies the aesthetic qualities and functional properties of the part, but it also restricts the selection of suitable manufacturing processes. Figure 2.7 is taken from Schey (1999) to show how one aspect of overall part geom- etry drives process choice.To quickly understand this graph, begin by noting that the cold rolling process nearest the x axis produces a flat strip that is at the extremes of wide and thin. In fact the strips are often several feet wide and still only a fewthou- sandths of an inch thick. No other process can match rolling for such dimensions. Cold rollingisthus one ofthe starting points for a large range of subsidiary processes such as sheet forming and stamping, which then produce automobile body panels, office furniture, and even the humble soup can. [...]... choice of an inherently expensive manufacturing process The next concept to emphasize is that of process chains within a particular family of manufacturing processes Examples of these are also shown on the Website In general, several processes are used sequentially to gradually achieve a highly accurate, smooth surface A common chain in mechanical manufacturing is to start with... the manufacturing line generate no sales • The total profit that starts to be gained at Tb• The key points on the time axis are: • Te~the project initiation point This is followed by product definition, product development, process planning for manufacturing, setting up machines, debugging the assembly line, and first launch into manufactured product around point Tm• • Tm-the point where real manufacturing. .. If the designer is unaware of these transitions, unnecessary finishing costs may be created, as shown in Figure 2.10 The other side of this coin is that manufacturing costs can he saved if the designer is willing to loosen desired tolerances • The manufacturing quality assurance at one step in the process chain must be carefully executed before moving on to the next process If a "parent" process is... will specify additional final finishing operations such as grinding and lapping These 38 Manufacturing Analysis: Some Basic Questions for a Start-Up Company 400 Figure2.10 part #" 300 i 200 ~ i ~ moves Chap 2 Finishingcostsincreaseasa from a rough casting.to a finish-machined part, to fine-honed final product (from Manufacturing Processes for Engineering Materials by Kalpakjian, © 1997 Reprinted by permission... product will run into years rather than months 2.3.10 Cost Factors Especially Related to Adjoining Parts The following example shows how design and manufacturing keep changing to suit a rather complex interaction between (a) the availability of innovative manufacturing techniques and (b) new economic conditions In other words, recalling Ayres and Miller's (1983) quotation in Chapter 1,"elM is the confluence... in the lower photograph This eliminates costly and unpredictable joining operations in the factory Thomas (1994) has observed that such manufacturing innovations flowing back into the design phase must be the new way of organizing the relationship between design and manufacturing Using Ayres and Miller's definition of CIM we can observe that the new innovations, or the new supply elements, include:... integrity of the aircraft 40 Manufacturing Analysis: Some Basic Questions for a Start-Up Company FIgure 2.11 component Chap 2 Integrated product and process design allows this aerospace to be completely machined from the solid as shown in the lower photograph (counesy of Dr Donald Sandstrom, The Boeing Company) These trends introduce a great deal of complexity into the design and manufacturing process, but... isolation There will always be something that can be improved, simplified, or made cheaper if design and manufacturing are viewed from a slightly wider system perspective 2.3.11 Analyzing Costs in Terms of the Profit Potential Hewlett-Packard's return map (RM) is another method for analyzing design and manufacturing costs However, it focuses not just on cost but on these survival questions: • How much... center of each process are the normally anticipated values This range is given the name natural tolerance (NT) of the process and is crucially important in both design and manufacturing work It cannot be emphasized enough that the cost of manufacturing, and the subsequent cost of any consumer product, is related to the designer's selection of part accuracy and dimensional tolerance Once the design and its... which measures (Tb ~ Tm) and focuses more on manufacturing productivity Obviously, fast production and high volumes of product are desirable The goal is to quickly amortize all the development costs • li.T and li.P-any arbitrary point (AT,AP) beyond the break-even-after-release time (Tm) where Hewlett-Packard's return factor (RF) is calculated The RFis 42 Manufacturing Analysis: Some Basic Questions . those who refined their design and manufacturing skills the best to give a high quality-to-cost ratio. Consumers are, Purchasing by real market adopters I 26 Manufacturing Analysis: Some Basic. all the costs involved in creating, manufacturing, and selling a product. At the bottom of the chart, the prime costs are for the direct labor and materials. Manufacturing overhead is then shown,. administrative Manufacturing charges Indirect materials Indirect labor Direct labor Direct materials Conversion costs Overhead =. 'PrUne cost. COSI of goods manufactured 'Cost of manufacturing, development.and