—汽产品开发流程

0 Product Development in the Automotive Industry: Strategies to Circumvent the Complexity Challenge Product Development in the Automotive Industry: Strategies to Circumvent the Complexity Challenge January 31, 2002 Ulrich Näher, Wolfgang Neubert, Arno Antlitz 1 4 5 6 7 8 9 10 11 12 5.0 7.5 10.0 12.5 15.0 Number of models is increasing and product life cycles are decreasing Source: Press clippings Models per brand Number Product life cycle Years 1987 2000 -30 %-30 % +25 %+25 % 2 60 24 60% Mondeo 1993/2000 Reduction driven by implemen- tation of simulation technologies Time-to-market is reduced dramatically * Start of production Source: Automobile production, AN, MID MONTH FROM DESIGN FREEZE TO SOP* 40 29 C-class 1995/2000 28%38 32 Golf III / IV 1991/96 16% Reduction driven by implementa- tion of stringent quality gates OUTSIDE-IN ANALYSIS 3 In addition, urgency towards innovation drives vehicle complexity Source: Automobil Entwicklung, survey results, McKinsey/ika Type of innovation in electronics Percent Radical innovation Radical innovation Individual innovation Incremental modification 100% Today 2010 50 30 20 27 30 43 • BMW Z22 carries 70 major innovations and 61 patents • Objective is to ensure new technology concepts for 2005 and beyond • Approx. 70 - 80% of innovations are in the field of electronics: – X-by-wire – Car PC – Center monitor – Fingerprint recognition – Head-up display Vehicle complexity – example BMW Z22 – Integrated starter/ alternator – Curvelight – Speech control – Cameras for rear view – Telematics Integration challenge 4 Key levers to address complexity challenge Clear and precise customer knowledge and orientation1 Efficient product architecture – from identity to similarity2 Value chain adaptation towards competence based structures3 Improved development processes leveraging IT opportunities4 Stringent quality processes along entire development process5 Project organization combining high functional and integration capabilities6 Source: McKinsey 5 27.2 14.6 10.2 1980 1990 2000 Increase in product variety and model change rate is driving passenger car market fragmentation Source: Schwacke 1998, Marketing Systems, EIU, Automobil revue, press clippings, McKinsey SHARE OF TOP-10 SELLING MODELS WESTERN EUROPE, 1980 - 2000 Percentage of total sales units 6 Market size Time Dimen- sion 1 Dimen- sion 2 B A C Competing vehicles A, B, C, D Market segments Market maturationMarket creation In mature and highly fragmented markets two strategies are possible: Targeting average vs. tailored market segment CONCEPTUAL Source: McKinsey Dimen- sion 1 Dimen- sion 2D A C B 1 43 2 OEM's model Dimen- sion 1 Dimen- sion 2D A C B 1 43 2 OEM's model Tailoring of models to spe- cific customer segments B Coverage of many promi- nent market segments A 7 To understand what customers really want is key Source: Automotive branding survey, May 2001 Stated Importance Out of 10 D e r i v e d I m p o r t a n c e Sporty Negotiations straightforward Agile Value customer Elegant Modern Attractive externally Support is good value ComfortableSuperior speeds Innovative Showroom experience positive For people in the know ExclusiveStands out A leader I feel attractive Proud to show this offMost highly regarded Exciting I look successful Won't let you down Efficient fuel usage Sufficient space Running costs reasonable Acceptable resale Information easy Around in 20 years Environmentally friendly An escape Youthful Manly Understated Won't break down Cost is good value Delivers what it promises Safe to drive For people like me 0.00 0.10 0.20 0.30 0.40 0.50 0.60 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 Attractive internally Fun to drive Customer want and state it Customer want but don't state it Customer state though really don't want it Derived importance dominated by emotional attributes Stated importance dominated by rational attributes 8 Key levers to address complexity challenge Clear and precise customer knowledge and orientation1 Efficient product architecture – from identity to similarity2 Value chain adaptation towards competence based structures3 Improved development processes leveraging IT opportunities4 Stringent quality processes along entire development process5 Project organization combining high functional and integration capabilities6 Source: McKinsey 9 Efficient architectures have to be optimized on vehicle and component level Focus of a standard- ized parts strategy Vehicle design (macroarchitecture) • Define packaging zones • Determine organization of components • Define levels of freedom for microarchitecture Ensure compatibility of macroarchitecture in family concepts (e.g., electronics architecture) Component design (microarchitecture) • Design components • Systematically optimize number of variants Increase share of standardized parts in vehicle family Organization of engine components Cable harness Interior dimensions, interior packaging Foot controls example Foot controls for vehicle type A Architecture redesign Foot controls for vehicle type B Joint component for vehicle family Source: McKinsey 10 Product architecture Parts/module architecture Identical parts/modules • 100% identical parts • Same variants across vehicle types • Building block modules Identity 100% 0% Adapted parts/modules Principle or concept parts/modules Solitary parts/modules • Existing parts/ modules with adjustments • Related functions or geometries ("pantograph") • Parts/modules specific to vehicle types Source: McKinsey Foot controls example Foot controls for vehicle type A Foot controls for vehicle type B Existing product architectures are redesigned with highest share of identical parts possible while maintaining sufficient differentiation Function separation Function integration Function elimination Variant combination Restructuring Combination reduction 11 Technology leaps P 400,000 800,000 Fixed-cost dilution P 400,000 800,000 Complexity cost reduction Potential levers Descrip- tion Examples • Payback on investments across large numbers of units • Reduction of variable costs by changing production concept • Lower flexibility requirements due to higher share of ongoing core operations • Reduction of variety costs • Much lower development costs for vehicle type B • Higher utilization of machinery • Increase in level of automation • Optimization of production site concept • Increased production on highly specialized, constantly running lines (fewer variants on one line) • Reduction of process costs at supplier: Purchasing, sales, production planning, administration, logistics, etc. Flexibility reserves reduction Units 200,000 400,000 2005 2010 2015 BR A & BR B BR A Increase in potential dependent on • Blocking type • Blocking level Source: McKinsey For deriving communality potentials four cost levers have to be understood 12 Fixed-cost dilution Technology leaps Flexibility reserves reduction Complexity reduction Bought-in materials Manufacturing costs Research and development Warranty and goodwill Administration and sales costs High impact Partial impact Low impact Impact at 100% com- munality Percent 7 - 9 9 - 10 10 10 5 Cost type EXAMPLE Source: McKinsey Cost types are impacted differently by cost levers Impact depends upon level of similarity/identity 13 Key levers to address complexity challenge Source: McKinsey Clear and precise customer knowledge and orientation1 Efficient product architecture – from identity to similarity2 Value chain adaptation towards competence based structures3 Improved development processes leveraging IT opportunities4 Stringent quality processes along entire development process5 Project organization combining high functional and integration capabilities6 Source: McKinsey 14 0 5,000 10,000 15,000 1993 1994 1995 1996 1997 1998 1999 2000 2001 Price increases above the inflation rate cannot be enforced despite new technologies Historical price development * Incl. value added tax ** Statistically not significant Source: HAWK project team List price VW Golf base model* EUR Not inflation-adjusted Inflation-adjusted Golf III Golf IV Additional charge potential for new technologies – example brake-by-wire Average additional costs for brake-by-wire Small car segment Compact car segment Medium car segment Luxury car segment Large car segment Additional charge potential 400 300 200 250 ~ 300 5,000 end customers surveyed CAGR 3.2% ~ 1,000** 15 4.000 3.000 11.000 15.000 12.000 Cost due to additional features have to be compensated by optimizing the value chain Source: HAWK project team Additional costs through new tech- nologies Car 2015 (with old industry structures) Synergy and enhancement processes Car 2015 best practice value chain architecture PRODUCTION COSTS COMPACT CAR, NOT INFLATION-ADJUSTED EUR/unit Car today Electronics share 20% Electronics share 40% ~ 20% cost effect through best practice value chain architecture and CIP CAGR 2.4% 16 INDUSTRY STRUCTURE Functional value chain architecture will give way to one that is know- how-driven Source: Expert interviews, HAWK project team In the futureToday System integration Elec- tronics Mechani- cal Division mainly by know-how because of • Economies of scale • Development synergies • Complexity Brake system Steering system Suspen- sion Axles Division mainly by function (system) or spatial placement (module) Brake system Steering system Suspen- sion Axles EXAMPLE: CHASSIS Functionality-/position-driven Know-how-driven Mechan- ical specialist Mechan- ical specialist Lenk- system- integra- tor Spring and shock absor- ber spe- cialist OEM Steer- ing system manu- facturer Brake system integra- tor System develop- ment OEM X-by-wire integrator Mechanical specialist Mechanical specialist Mecha- tronics specialist Mechanical integrator Brake system inte- gra- tor 17 Specific competencies are required to capture new synergies Source: Team HAWK NEW SYNERGY POTENTIAL IN CHASSIS SEGMENT USD per vehicle 130 38 31 199 Synergy potential through value chain optimization Synergy potential for x-by- wire integrator Synergy potential for mechanical specialists Synergy potential for OEM Synergies Required competencies EXAMPLE CHASSIS X-by-wire-integrator • Centralization of ECUs and basis software • Scale effects and optimization of interfaces between new electronic components • Economics of scope in production of sensors and actuators • System integration (e.g., ECU centralization) • Innovative creativity (e.g., ECU and Software design) • Development efficiency in electronics (e.g., sensors) • Operational excellence (e.g., actuators, sensors) Mechanical specialists • Economics of specialization for mechanical components • Operational excellence • Ability to capture scale effects • Factor cost efficiency OEM • Avoidance of interfaces through centralized chassis control via software • Transaction cost efficiency • Understanding of customer needs 18 Detailed analysis of competency gaps helps to derive specific activities • Competency building is needed, particularly in the areas of development efficiency for electronics and innovative drive • Competency gap could be closed by means of cooperating with an innovative electronics specialist Source: HAWK project team Competencies Mechanical development efficiency Best-practice company Sample company Competency gap 0 6 COST REDUCTION POTENTIAL FOR FUTURE STEERING SYSTEM INTEGRATOR Percent Realizing operational excel- lence/economies of scale Electronics development efficiency Innovative drive Module/system integration Factor cost efficiency Overhead/transaction cost efficiency Purchasing efficiency Understanding of end customer EXAMPLE: CHASSIS 19 Key levers to address complexity challenge Clear and precise customer knowledge and orientation1 Efficient product architecture – from identity to similarity2 Value chain adaptation towards competence based structures3 Improved development processes leveraging IT opportunities4 Stringent quality processes along entire development process5 Project organization combining high functional and integration capabilities6 Source: McKinsey 20 -50% 0 20 40 60 80 1985 1990 1995 2000 2005 The 2005 target requires a reduction of development times by 50% * Concept-freeze to SOP Source: Publications on vehicle development times (70 vehicles worldwide) between 1988 and 2000, McKinsey-Research 60 months (1988) 35 months* (1999) 42 months (1991) 40 months (1994) Target 30 months or less AVERAGE DEVELOPMENT TIMES, PROJECT DECISION TO SOP Month 21 A near future development process is characterized by virtual techniques and only 1 prototype cycle 30 MONTH DEVELOPMENT PROCESS Month Steps Gateways Package Styling -35 -30 -23 0 Start of project Concept decision Start of production Package Engineering/ CAE Prototypes Testing Production test series Exterior/Interior Design Design cycles CAD 100% (-17) Prototype cycle Component tests Integration tests Endurance tests Validation Industriali- zation Concept development Series development/-preparation Project planning Design freeze (-23) Massive use of virtual simulation Package definition (-23) -5 Package freeze (-19) Virtual steps/process development Ramp- up Pre-series tests Source: Harvard Business Review Optimized test strategy driving cross functional vehicle perspective One prototype cycle for critical vali- dation tests only 22 Product testing must be optimized along different dimensions Effective concept Efficient execution Impact • Specific parameters can be tested very early • Test of more variants/options due to faster test cycles • Significant reduction of effort • Early test of highly critical criteria/properties • Cost reduction • Test planning – Risk prioritization – Optimization of utilization – Cross-functional use of prototypes • Execution of tests – Automation – Up-Speeding Complete product System Compo- nent Simu- lation Labo- ratory Field test 1 2 Source: McKinsey AUTOMOTIVE EXAMPLE 23 Key levers to address complexity challenge Clear and precise customer knowledge and orientation1 Efficient product architecture – from identity to similarity2 Value chain adaptation towards competence based structures3 Improved development processes leveraging IT opportunities4 Stringent quality processes along entire development process5 Project organization combining high functional and integration capabilities6 Source: McKinsey 24 Reduced profitability through potential problems at ramp-up/ SOP Forgone sales Increased cost Possible SOP problems (assumptions) Opportunity potential* USD millions** Late market launch Reduced production capacity Target production cost exceeded Resources used for ramp-up/SOP Warranty and goodwill cost Cost of changes Target development cost exceeded Market launch 6 months late Customer migration 15 percent migration of former customers Full production reached 6 months late 10 percent over target production cost 15 percent over target development cost 50 percent over target SOP cost Long-term quality problems Ø USD 400/vehicle Changes to body pressing tools 6 months before SOP Maturity problems at ramp-up/SOP have significant impact on profitability * Profit contribution from profits or cost differences over life cycle, assuming: 500,000 units p.a., USD 5,000 profit contribution/vehicle, production time 7 years ** Over total production time Source:McKinsey ~750 ~1,250 ~500 ~2,000 ~190 ~125 ~1,400 ~250 25 Share of electronics and software problems 10.6 11.3 12.7 13.2 15.3 17.7 18.4 19.2 19.9 41.9 42.4 66.4 52.8 Saab Alfa Daewoo Fiat Mazda Subaru Honda Toyota BMW Nissan VW Porsche Audi 53% 48% 49% 55% 44% 53% 48% 46% 46% 45% 48% 55% 44% Failures per 1000 vehicles thereof caused by electronics and SW SW problems are reasons for recall of more than 700.000 vehicles in 2002 Software maturity is becoming a critical factor in automotive product development Software-related Source: McKinsey, Business Week, ADAC-AutoMarxX (3-5 year old car failures 1998-2001), cars in Germany only Source of quality problems Malfunction in Percent Infotainment and body electronics Injection/ignition system Engine (w/o injection) Radiator/cooling Wheels/tires Fuel systems Other Gears/transmission 5 4 6 6 7 8 12 20 32 Chassis 26 Automotive software development adds a new layer of complexity compared to hardware Source: Brooks: The Mythical Man-Month, McKinsey • High number of tacit requirements • Heavy software and hardware interaction for embedded systems • Project complexity growing steeply with product size • Intangible product, hard visualization and performance tracking • General mismatch between scope and available resources - projects always seem to be "nearly" complete More complexity Less trans- parency Fundamental differences: Find specific solution Fundamental differences: Find specific solution • High degree of change in underlying complex technologies • No widely accepted platform standards • Immature tool landscape • Fast-moving (and in many cases immature) markets • Customer value hard to assess • Lack of experience translating customer requirements into functionality • Inherent tendency to over-engineering • Seemingly low cost of changes • Invariant resource under-estimation • Irrational developer preferences Less discipline More technolo- gical risk More business risk Will disappear as industry matures: Learn from hardware Will disappear as industry matures: Learn from hardware 27 Source: McKinsey Development organization Product architecture Process efficiency Operational improvement can be achieved in a three step approach • Restructuring of development organization for specific needs of SW projects is necessary • Building of specific skills in SW development and SW project management is needed • Modular, feature specific product design is key to reduce complexity and enable concurrent engineering • Platforming and maximal degree of reuse is neces- sary to overcome complexity challenge and ensure software quality • Complex software projects are only feasible with standardized, repeatable processes • Development effort depends heavily on process maturity - efficiency potentials of up to 90% are possible 28 Automotive industry C h a r a c t e r i s t i c s C M M l e v e l Source: McKinsey Disciplined process Initial • Undefined pro- cesses, ad hoc working methods • Success de- pends on few specialists • Schedule, quality and cost unfore- seeable Repeatable • Process owned by project manager • Disciplined project management • Process varies from project to project Defined • Standard pro- cess owned by organization • Process-specific tailoring of the standard process Managed • Quantitative goals for product and process • Tracking of goals by metrics and statistical analysis Optimizing • Process change management • Defect prevention processes • Technology charge management Standard consistent process Predictable process Continuously improving process Automotive industry target !!!!Aircraftin