me579 14 Platform optimization

Slide1: 

PENNSTATE Timothy W. Simpson Associate Professor Mechanical & Nuclear Engineering and Industrial & Manufacturing Engineering The Pennsylvania State University University Park, PA 16802 phone: (814) 863-7136 email: tws8@psu.edu http://www.mne.psu.edu/simpson/courses/me579 ME 579 - Designing Product Families - IE 579 © T. W. SIMPSON *These slides are adapted, with permission, from Prof. Olivier de Weck at MIT.

Impacts on Platform Strategy: 

Market Segmentation Mass Customization Platform Strategy Manufacturing & Supply Chain Strategy Product Architecture Regulations & Standards Customers: Perceived Value Product Design Competition: New Products New Technologies Impacts on Platform Strategy

Need for Platform Strategy: 

Need for Platform Strategy Competition: How to preempt or react quickly to new products from competitors? Customers: What product features do all customers value highly? What product features are requested infrequently? Technology: Can a product platform be designed such that new technologies can be “easily” infused? Regulations and Standards: Can a platform be design to anticipate or meet future regulations (e.g. fuel economy and emission standards in cars)? Strategy: An adaptation or complex of adaptations (as of behavior, metabolism, or structure) that serves or appears to serve an important function in achieving evolutionary success. (Merriam-Webster, 2004)

Strategic Framework: 

Product Platforms Production Plants (Facilities) Variants Models Market Segments (Customers) Platform Family Plan Manufacturing Plan Marketing Plan to maps maps to assigns Platform A Platform B Platform C Chevrolet Malibu Saturn SL determines GMC Truck Sierra 1500 places chooses in SUV PICKUP SEDAN PLANT A PLANT B PLANT C Corporate Strategy Domain of Product Platform Strategic Framework

Product Architecture Business Strategy: 

Product Architecture Business Strategy Product Portfolio or Collection of Products? Open Systems? Platforms? or Cost-optimized Products Organized to support architectural directions? (More than just an engineering question) Does the system architecture match the evolution of key emerging technologies? (e.g., the Internet?) Does the system architecture match the evolution of key surround business changes? (e.g., competing on cost or function? Changing product distribution channels?)

Starting Assertions: 

Starting Assertions Maximize profits …how? … offer family of diverse, competitive product variants, and minimize mfg cost Platform strategy = program of deliberate reuse of components and processes within a family How can commonality between products be quantified?  commonality indices What components should be shared between products?  expensive ones with little effect on variant distinctiveness What is the optimum amount of commonality?  difficult to answer in general, depends on market, firm …  moving target, changes dynamically from year-to-year Without competition, no need for variants, no need for platforms  Ford Model T (one size fits all)

Platform Strategies: 

Platform Strategies Usually start with a market segmentation grid Low-End Mid-Range High-End Luxury Brand A Brand B Brand C Brand D “Market Segment” Vertical Leveraging Beachhead Approach No Leveraging Horizontal Leveraging

Platform Portfolio Problem: 

Platform Portfolio Problem How many platforms (N) are optimal to support (V) variants?  Optimize Ratio V/N What is the optimal assignment of the V variants to the N platforms?  Optimize assignment How to deploy the V variants across M target market segments?  Optimize Market Segment assignment Determines Platform “Extent” Platform Portfolio Product Family a b g A Y Z Y Z C GM N~20 V~100 For large product families, need more than one platform Ref also: Seepersad, Mistree, Allen, “A quantitative approach to designing multiple product platforms for an evolving portfolio of products”, 2002 ASME Design Engineering Technical Conferences, Paper No. DETC2002/DAC-34096

Profit Maximization: 

Profit Maximization Maximize product family profit, subject to investment cost constraints, by determining the optimal - number of platforms N - assignment vector - platform design vector set - variant design vector set revenue cost Sales volume Actual Sales Price Total Cost Sum over all V product variants de Weck O., Suh E. S. and Chang D., ”Product Family and Platform Portfolio Optimization”, Paper DETC03/DAC-48721, Proceedings of DETC’03, 2003 ASME Design Engineering Technical Conferences, Chicago, Illinois, 2-6 September, 2003

Product (Variant) Modeling Framework: 

Product (Variant) Modeling Framework Engineering Model Architecture Model Manufacturing Model Value Model Market Model Financial Model J performance V value P D demand price Pr net profit x C cost design vector c components maximize requires 6 models

Bi-Level Optimization Methodology: 

Bi-Level Optimization Methodology Variant Model Variant Optimizer Platform Optimizer Platform Model optimize for each N=1,2,…V Pr Steps Create 6 models Split product architecture into platform and variant components Select N=1 platforms Perform bi-level optimization for Pr Set N=N+1 if N<V Repeat steps 4. and 5. until N=V

Automotive Case Study (hypothetical): 

Automotive Case Study (hypothetical) What is the “optimal” platform strategy for a family with 7 variants ? One product (vehicle) per market segment Basic vehicle architecture is always BOF Market segments operate independently Competitors continue to offer the same MSRP corresponds to actual sales price Use 2001 database for North America The platform consists of the chassis Assumptions low mid high How many platforms? How to optimally assign variants to those platforms? Market Segmentation Grid

Vehicle Data Set for Case Study: 

Vehicle Data Set for Case Study We are a (new) automotive manufacturer and want to compete successfully in these market segments: Symbol Name #vhc Size Mean Price LOWC Compact Car 30 2,357,802 $13,427 MDSD Medium Sedan 33 4,198,028 $19,844 LXSD Luxury Sedan 65 1,591,438 $34,238 SPTR Sports-Roadster 34 514,837 $23,424 SUVC SUV 56 3,519,461 $25,146 PUPT Truck 51 2,800,104 $22,805 MVAN Van 24 1,589,958 $24,986 16,571,628 Total U.S. Market 2001 ca. 16.8 M/year New Vehicle Sales Source: NIADA National Market Report - 2002 What is the right strategy?

Architecture Model: 

HT body Architecture Model WB WT ED chassis (common) engine vehicle Design Vector Architecture Chassis Engine Body

Vehicle Design Vector: 

Vehicle Design Vector (genotype = vehicle “DNA”) DV=[ WB WT ED HT SF ]T Example: DV=[ 108.2 61.3 2990 58 1.0 ]T Platform Engine Body PDV(k) =[ WB WT]T MDV(j) =[ ED ] DV’ = [ k j 1400 1.0 ] DV’’=[ 0 1 1 | 1 1 1 | 1 0 1 0 0 1 1 1 | 0 1 1 1 0 1 0 1 ] encode decode Units [in] [in] [ccm] [in] [-] binary

Engineering Model (e.g. WB to FE): 

Engineering Model (e.g. WB to FE) Instead of detailed CAD/CAE-simulation model: - Response Surface Modeling (RSM) - Neural Network Regression Models x(1)=WB J(3)=FE

Value Model: 

Value Model AC - Acceleration HP - HorsePower FE - Fuel Economy PV - Passenger Vol. CV - Cargo Volume LOWC MDSD LXSD SPTR SUV TRCK VAN 0.1 0.15 0.15 0.4 0.1 0.15 0.05 0.1 0.1 0.15 0.3 0.25 0.35 0.1 0.4 0.2 0.05 0.05 0.05 0.10 0.05 0.3 0.4 0.45 0.2 0.3 0.05 0.4 0.1 0.15 0.2 0.05 0.3 0.35 0.4 Preference weight matrix (Perceived) Value = Aggregate performance relative to the market segment leader Value= Relative Performance Relative Price Performance Vector J Attributes

Example Segments: 

Example Segments Compact Cars -Sales vs MSRP $0 $5,000 $10,000 $15,000 $20,000 $25,000 0 100000 200000 300000 400000 Sales Volume MSRP Sales Volume vs MSRP - SUVs $10,000 $20,000 $30,000 $40,000 $50,000 $60,000 $70,000 $80,000 0 100000 200000 300000 400000 500000 Sales Volume MSRP Compact Cars - LOWC Sports Utility Vehicles - SUV Who are the leaders? Source: AutoPro

“Sweet Spot” - Market Model: 

“Sweet Spot” - Market Model Relative Position w.r.t Leader - LOWC $0.60 $0.70 $0.80 $0.90 $1.00 $1.10 $1.20 $1.30 $1.40 $1.50 $1.60 0.800 0.900 1.000 1.100 1.200 Value = Relative Performance Relative Price Prel Dw,i

Demand Sensitivity Curve: 

Demand Sensitivity Curve Demand Sensitivity - MDSD 0 50000 100000 150000 200000 250000 300000 350000 400000 450000 0.00 0.20 0.40 0.60 0.80 1.00 Weighted Distance from Leader Dw Demand = Sales Volume Using common components (e.g. platforms) reduces design freedom Reduced design freedom increases distance from the “sweet spot” Sales Volume (Demand) drops as we increase the sweet spot distance Demand Sensitivity Curve quantifies penalty due to platforming

Cost (Manufacturing) Model: 

Cost (Manufacturing) Model Market Leader MSRP Vehicle Cost Frame Engine Body 45% 25% 30% - x% Include Learning Curve Effect Total Product Family Cost: Margins x%: LOWC 5% MDSD 10% LXSD 20% SPTR 15% SUV 15% Truck 25% Van 15% 100% 100-x%

Simulation/Optimization Framework: 

Simulation/Optimization Framework i=1,..,7 nnet opt 1 opt 2 mkt cost platform # of platforms portfolio Vehicle Level Family Level profit Bi-Level Optimization Framework

Resulting “Optimal” Platform Strategy: 

Resulting “Optimal” Platform Strategy LOWC MDSD LXSD SPTR SUV VAN Utility Sedans Sports TRCK a 1 1 1 1 1 1 1 a,b 2 2 2 2 2 1 2 a,b,c 3 3 2 3 2 1 2 a,b,c,d 4 3 2 4 2 1 2 a,b,c,d,e 4 3 2 4 2 1 5 a,b,c,d,e,f 4 3 2 6 2 1 5 a,b,c,d,e,f,g 4 3 7 6 2 1 5 Increasing # of platforms “Optimal” variant- platform assignment matrix V N

Platform Strategy Evolution: 

Vehicle X1 Vehicle X2 Vehicle X3 Vehicle Y1 Vehicle A4 Vehicle C2 Vehicle Platforms Engine Suspension a b c d e f ei1 ei2 ei3 ev2 ev3 Platforms BOM Current Vehicle Family sfl smp srs ... Newly proposed vehicle What is the “best fit” existing platform for this new vehicle ? ? Platform consolidation proposal ? ? What are the consequences of consolidating a platform? Platform Strategy Evolution