10241_9789813148253_tp.indd 22/11/16 2:46 PM WSPC Series in Advanced Integration and Packaging Series Editors: Avram Bar-Cohen (University of Maryland, USA) Shi-Wei Ricky Lee (Hong Kong University of Science and Technology, ROC) Published Vol 1: Cost Analysis of Electronic Systems by Peter Sandborn Vol 2: Design and Modeling for 3D ICs and Interposers by Madhavan Swaminathan and Ki Jin Han Vol 3: Cooling of Microelectronic and Nanoelectronic Equipment: Advances and Emerging Research edited by Madhusudan Iyengar, Karl J L Geisler and Bahgat Sammakia Vol 4: Cost Analysis of Electronic Systems (Second Edition) by Peter Sandborn Chelsea - Cost Analysis of Electronic Systems.indd 02-08-16 10:43:54 AM 10241_9789813148253_tp.indd 22/11/16 2:46 PM Published by World Scientific Publishing Co Pte Ltd Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library WSPC Series in Advanced Integration and Packaging — Vol COST  A NALYSIS  OF  ELECTRONIC  SYSTEMS Second Edition Copyright © 2017 by World Scientific Publishing Co Pte Ltd All rights reserved This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the publisher For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA In this case permission to photocopy is not required from the publisher ISBN 978-981-3148-25-3 Printed in Singapore Chelsea - Cost Analysis of Electronic Systems.indd 02-08-16 10:43:54 AM Preface to the Second Edition I received helpful criticism from numerous sources since the first edition of this book was published in 2013 In addition to the first edition’s use as a graduate course text, we are now using selected chapters in an undergraduate course on engineering economics and cost modeling Along with the inputs I have received on how to make the original topics more complete, I have also had numerous requests for new material addressing new areas Of course no book like this can ever be truly complete, but attempting to make it so keeps me out of trouble and gives me something to on the weekends and evenings I have added two new chapters and two new appendices to this edition The new chapter on real option analysis treats modeling of management flexibility and provides a case study on maintenance optimization A chapter on cost-benefit analysis has also been added This chapter comes as the direct result of many inquiries about how to model consequences (benefits, risks, etc.) concurrent with costs The new appendices cover weighted average cost of capital and discrete-event simulation, both of these topics don’t warrant a chapter, but nonetheless are useful topics for this type of book In addition to the new chapters and appendices, several new sections have been added to the 1st edition chapters and new problems have been added to all the chapters (and a few problems that students convinced me didn’t quite make sense have been deleted) Peter Sandborn 2016 v b2530   International Strategic Relations and China’s National Security: World at the Crossroads This page intentionally left blank b2530_FM.indd 01-Sep-16 11:03:06 AM Preface to the First Edition Twenty years ago many engineers involved in the design of electronic systems took, at most, a secondary interest in the cost effectiveness of their design decisions; they considered that someone else’s job or an issue to be addressed after the initial release of the product.1 Today the world has changed Every engineer in the design process for an electronic product is also tasked with understanding, or contributing to the understanding of, the economic tradeoffs associated with their decisions Yet aside from general engineering economics that focuses on capital allocation problems, system designers have virtually no resources and obtain little or no training in cost analysis, let alone analysis that is specific to electronic systems Unfortunately, when engineering students were asked what they thought the cost of a product was (and assigned to determine cost estimates of products in an undergraduate capstone design course at the University of Maryland) they all too often added up the costs of procuring the bill of materials and declared that to be the cost of the product Few students are surprised when shown a breakdown of the life-cycle costs or the cost of ownership of systems, but virtually none, even those who had taken courses in engineering economics, were equipped to competently estimate the manufacturing or life-cycle cost of a real product This book is an outgrowth of a course on Electronic Product and System Cost Analysis developed at the University of Maryland Since 1999, the course has been taught as a one-semester graduate course (populated with a mix of senior-level undergraduates and graduate students) and many times in the form of an industry short course Many types of electronic systems have been primarily driven by time to market rather than cost; this situation is not necessarily shared by non-electronic systems vii viii Cost Analysis of Electronic Systems This book is intended to be a resource for electronic system designers who want to be able to assess the economic impact of their design decisions on the manufacturing of a system and its life cycle The book is oriented toward those interested in the entire electronic systems hierarchy from the bare die (integrated circuits) through the single chip packages, modules, boards, and enclosures This book provides an in-depth understanding of the process of predicting the cost of systems Elements of traditional engineering economics are melded with manufacturing process modeling and lifecycle cost management concepts to form a practical foundation for predicting the real cost of electronic products Various manufacturing cost analysis methods are included in the book: process-flow cost modeling and parametric, cost-of-ownership, and activity-based costing The effects of learning curves, data uncertainty, test and rework processes, and defects are considered in conjunction with these methodologies In addition to manufacturing processes, the product lifecycle costs associated with the sustainment of systems are also addressed through a treatment of the cost impacts of reliability (sparing, availability, warranty) and obsolescence The chapters use real-life scenarios from integrated circuit fabrication, electronic systems assembly, substrate fabrication, and electronic systems testing and support at various levels The chapters contain problems of varying levels of difficulty, ranging from alternative numerical values that can be used in the examples included in the chapter text to derivations of relations presented in the text and extensions of the models described Even for the simple problems, students may have to reproduce (via spreadsheet or other methods) the examples from the text before attempting the problems The notation (symbols) used in each chapter are summarized in the Appendix Every attempt has been made to make the notation consistent from chapter to chapter; however, some common symbols have different meanings in different chapters The author is grateful to many people who have made this a much better book with their input First, I want to thank the several hundred students who have taken courses at the University of Maryland and seem to somehow always find new and unique questions to ask every time it is taught My graduate students, present and past, deserve appreciation for Preface to the First Edition ix their contributions to many portions of the book In particular I would like to acknowledge Andre Kleyner (Delphi) and Linda Newnes (University of Bath) for their contributions reading and commenting on several of the chapters I would also like to thank my numerous colleagues at the University of Maryland and in CALCE, including Michael Pecht and Avi Bar-Cohen for encouraging the writing of this book Peter Sandborn 2013 b2530   International Strategic Relations and China’s National Security: World at the Crossroads This page intentionally left blank b2530_FM.indd 01-Sep-16 11:03:06 AM 540 Cost Analysis of Electronic Systems the timeline has to be modeled sequentially for all events An example of this would be multiple system instances drawing spares from a common inventory In this case, separate DESs for each system instance cannot be generated and then added because the timing of spare replenishment (which represents an event that costs money) depends on the demands from all of the system instances In this case, both system instances have to be simulated concurrently Discrete-event simulators also suffer from the constraint that they only operate in one direction, i.e., forward in time Because of this, there are many outputs of discrete-event simulators that are straightforward to generate (e.g., cost and availability) that become very difficult to use as inputs to a design process For example, availability requirements can be satisfied by running discrete-event simulators in the forward direction (forward in time) for many permutations of the system parameters and then selecting the inputs that generate the required availability output Such “brute force” search-based approaches are computationally impractical for real problems (particularly for real-time problems), and are unable to deal with general uncertainties There have been attempts to perform reverse simulation (run discrete-event simulators backwards in time) but this has only been demonstrated on extremely simple problems with limited applicability to the real world systems References C.1 C.2 Nance, R E (1993) A History of Discrete Event Simulation Programming Languages, TR 93-21, Virginia Polytechnic Institute and State University, Department of Computer Science Conway, R W., Johnson, B M and Maxwell, W L (1959) Some problems in digital systems simulation, Management Science, 6(1), pp 92-110 Discrete-Event Simulation (DES) 541 Bibliography In addition to the sources referenced in this chapter, there are many books and other good sources of information on discrete-event simulation, including: Banks, J., Carson II, J S., Nelson, B L., and Nicol, D M., (2009) Discrete-Event System Simulation, 5th Edition, Prentice Hall Leemis, L M., and Park, S K., (2006) Discrete-Event Simulation: A First Course, Person Prentice Hall Problems C.1 C.2 C.3 C.4 C.5 C.6 C.7 In the simple example in Section C.2.1, several implicit assumptions were made about when failures occur and how they have to be fixed Identify and discuss these assumptions Rework the example in Section C.2.2 assuming that the time to failure is given by a Weibull distribution with the following parameters: location parameter = 500 hours, shape parameter = 4, and the scale parameter = 10,000 hours Rework the example in Section C.2.2 (with the constant failure rate), assuming that the time to resolve the failures (which was previously assumed to be instantaneous) is given by a triangular distribution with a lower bound of 30 days, an upper bound of 60 days and a mode of 45 days Is the cumulative cost larger or smaller than the cumulative cost when the failures are resolved instantaneously? Calculate the final (after 20 years) time-based availability of the system in Problem C.3 What if an infrastructure charge of $150/month is incurred in the example in Section C.2.2 (with the constant failure rate)? What is the total cost after 20 years? Hint: the infrastructure charge represents an event that is independent of the maintenance events Starting with the example in Section C.2.2 (with the constant failure rate), assume that each maintenance event requires one spare For simplicity, assume that the spare costs $1000 and the spare is the only maintenance cost – this is effectively identical to the solution in Section C.2.2 Now assume that the spares are kept in an inventory and that the inventory initially has spares in it (purchased for $1000 each at time 0) Whenever the inventory drops below spares, more replenishment spares are ordered (for $1000 each) Assume that the replenishment spares arrive instantaneously What is the total cost after 20 years? Suppose that the time-to-failure distribution used in the simulation in Section C.2.2 was for a particular part in a system and that the part becomes obsolete (nonprocurable) at the instant the simulation begins If you had to make a lifetime buy of parts to support this system through 20 years, how many would you buy? Index arbitrage, 483 artificial neural network, 105 asymmetric problem, 360, 482 automatic test equipment, 141, 149 availability, 242, 270, 325 achieved, 329 availability factor, 344 average, 328 computation, 332 contracting, 344 definition, 325, 326 energy-based, 343 Erlang-B, 341 example, 334 inherent, 328 instantaneous, 326 intrinsic, 329 joint, 332 Markov models, 336 materiel, 342 mission, 331 Monte Carlo example, 334, 335 operational, 329 optimization, 351 parallel systems, 349 random request, 332 series systems, 348 spares demand driven, 338 steady-state, 328, 341 supply, 330, 339 time-based, 325 unavailability, 274, 349 acceleration factor, 315 accounting, 1, 80 accuracy, 4, 15 absolute, relative, 4, 461 acquisition reform, 357 active inventory, 343 activity-based costing, 72, 77, 80, 375 activities, 80 activity base, 81 activity cost pool, 81 activity rate, 81 applicability to cost modeling, 79 concept, 78 cost objects, 80 example, 82 formulation, 79 history, 78 overhead allocation, 81 transactional drivers, 81 activity-based management, 78 administrative delay time, 329 advanced electronic power systems module, 174 after-tax, 247 Airbus, 95 airliners, 287 analytical models, 15 anomaly detection, 395 Apple 128GB iPhone 6+, 17 application specific integrated circuits, 97 543 544 Cost Analysis of Electronic Systems work-mission, 331 availability-based contracting, 344, 406 outcome-based contracts, 344 performance-based logistics, 347 power purchase agreements, 346 product service systems, 346 public-private partnerships, 347 backorders, 339 base rate fallacy, 473 benefit, 450 direct tangible, 451 indirect tangible, 451 intangible, 450 benefit-cost analysis, see cost-benefit analysis Bernoulli trials, 124 Beta distribution, 335, 362 bid, bill of materials, bin, 37 binomial coefficient, 39, 125, 349 distribution, 39 probability mass function, 124 series, 39 Boeing, 95, 104 Boeing 737, 394 learning curve model, 213 bottleneck, 142 bottom-up, 19, 94, 422 Buffon’s needle, 188 built-in self test, 143 burden rate, 10, 65, 80 burn-in, 259, 313 cost, 314, 315 definition, 313 example, 318 life removed, 316 manufacturing cost, 321 repairable units, 322 return on investment, 318 test time, 315 value, 317 business case, 245, 460 cannibalization, 340 capability indices, 54 capacity, 23, 26 capital allocation, capital costs, 9, 64 carrying cost, see holding cost cash flow, 247 central limit theorem, 197 certification, 262 change-over, 66 hot, 70 chi-square test, 195 circuit sensitivities, 227 classification model, 467 binary classifier, 467 class definition, 466 false positive, 468 majority-class event, 466 minority-class event, 466 threshold, 468 true positive, 468 unbalanced classes, 466 clustering parameter, 46 COCOMO, 419 COCOMO II, 422, 427 embedded model, 420 organic model, 420 semi-detached model, 420 coefficient of determination, 102 commercial off-the-shelf, 357 conceptual design, condition-based maintenance, 391 confidence interval, 197 confidence level, 277 consequence, 461 conservation of defects, 116 continuous compounding, 302 continuous improvement, 113 Index contract, 287 conversion matrix, 116 convolution theorem, 294 cool down, 65 correlation coefficient, 222 cost analysis, definition, cost avoidance, 244, 369 return on investment, 391 cost-benefit analysis, 449 benefit-cost ratio, 455 double counting, 459 example, 451 flaws, 459 netted-out, 455 value of human life, 456 cost effectiveness analysis, 460 cost estimating relationships, 93, 407 bounds of the data, 100 forced correlation, 103 historical data, 103 limitations, 100 overfitting, 101 scope of the data, 101 cost of doing nothing, 456 cost of money, 281 cost of ownership, 61 algorithm, 62 capital costs, 64 comparison of two machines, 67 definition, 61 modeling, 64 performance costs, 66 product costs, 71 sustainment costs, 64 cost of the status quo, 456 cost savings, 391 costing by analogy, 106 counterfeit parts, 358 customer, 5, 242 customer satisfaction value, 317 545 cycle time, 23 dead time, 142 debugging, 210 decision tree analysis, 477, 479 defects, 29, 35, 36 accumulating, 46, 47 clustering, 129 conservation, 116 coverage, 120 definition, 114 density, 37, 41, 43, 145 fatal, 36, 37 gross, 36 latent, 37 level, 127 non-fatal, 36 non-repairable, 66 parametric, 36 random, 37 relation to faults, 115 repairable, 66 spectrum, 115, 116, 120 Defense Procurement Reform Act, 288 Dell Computer, 262 demand forecasting, 359 dependent variable, 96 depot, 341 depreciation, 9, 28, 64 depreciation life, 25, 146 design, design for test, 140, 143 design refresh, 369 definition, 369 design refresh planning, 378 MOCA model, 373 Porter model, 369 device under test, 115 diagnosis, 113, 155, 156 definition, 114, 156 depth, 157 diagnostic length, 157 546 Cost Analysis of Electronic Systems diagnostic resolution, 157 diagnostic test, 156 diagnostic tree, 157 die, 27, 38, 98, 141 tiling fraction, 146 diminishing manufacturing sources and material shortages, see obsolescence Dirac delta function, 42 direct costs, see recurring costs discount factor, 246, 454 discount rate, 246 discounted cash flow analysis, 387, 477 discrete compounding, 247 discrete-event simulation, 373, 477 disruptive technologies, 104 downtime, 326 DuPont, 381 echelon (single and multi), 341 economic order quantity, 279, 372 economic production quantity, 279 edge scrap, 27 effectiveness, 350 electronic parts, 437 assembly model, 440 field failure model, 441 obsolescence, 357 part site, 440 part support model, 438 electronic signaling system, 451 embedded resistors, 74 emulation, 15 end of life, 5, 361 end-of-period convention, 246 Energy Star program, 264 EnergyGuide labels, 239 engineering change orders, 78 engineering economics, 2, 246 Environmental Protection Agency, 262 EPROM, 103 equipment and facilities-centric products, 18, 62 equipment costs, see capital costs Ericsson AB, 443 Erlang-B, 341 arrival rate, 342 blocking probability, 341 Erlang, 342 traffic intensity, 341 error, 114 escape fraction, 132 exponential distribution, 259 F-16 aircraft, 241 failure, 114, 252 avoidance, 251 cumulative, 254 distributions, 256 mechanisms, 464 misuse, 253 overstress, 253 rate, 258 wear-out, 252, 253 failure in time, 444 failure modes and effects analysis, 465 cost-based FMEA, 465 scenario-based FMEA, 465 fallout, 20 false positive, 133, 156, 163, 173 test step, 135 false positive paradox, 471 fault, 36 coverage, 120, 122, 160 coverage relation to yield, 122 definition, 114 dictionary, 157, 158 efficiency, 120 isolation, see diagnosis probability, 37 relation to defects, 115 simulation, 121 spectrum, 115, 116 type, 11 Index feature points, 423, 426 feature-based costing, 104 Federal Aviation Administration, 262 Federal Communications Commission, 262 fighter jets, 94 final-order problem, see obsolescence, lifetime buy fixed cost, see non-recurring cost fleet, 340 flip chip bonding, 151, 385 Food and Drug Administration, 262 footprint, 242 gate count, 97 General Electric, 78 geometric Brownian motion, 493 good-as-new, 260, 292 good-as-new repair, 291 half Gaussian distribution, 45 hazard rate, 257, 258 hazardous waste disposal costs, 110 heuristic models, 15 hidden costs, 10 hierarchy (of modeling), 15 high specification limit, 55 holding cost, 279, 360, 370 hypergeometric distribution, 125 IBM, 385 inactive inventory, 343 incentives, 210 independent variables, 96 indirect costs, see overhead costs inflation, 248 inflation rate, 248 market discount rate, 248 nominal method, 248 real method, 248 ink jet printer, 239 inner-layer pairs, 74 547 inner-lead bond pads, 151 inspections, see test integrated circuit, 26, 29, 140, 386 Intel Corporation, 61 intensity function, 305 inter-arrival time, 66 interest rate, 246 International Technology Roadmap for Semiconductors, 46 inter-occurrence times, 293 inventory, 339 inventory lead times, 282 inventory model, 271 inventory obsolescence, 281, 355 ISO 8402:1986, 35 ISO certification, 387 iterative, 189 kerf, 27, 98 kit, 275 known good die, 48 Kronecker delta, 42 labor burden, 10 cost, 8, 10, 23 rate, 10, 23, 65 labor-dominated products, 17 Latin hypercube, 200 Latin hypercube sample, 201 layer pair, see inner-layer pairs leadframe, 116, 386 learning curves, 209 algebraic midpoint, 219 block data, 224 Boeing model, 213 comparing learning curves, 220 Crawford model, 213 cumulative average learning curve, 213 De Jong model, 212 defect density learning, 231 548 Cost Analysis of Electronic Systems definition, 209 determining from actual data, 222 history, 209 learning index, 211 learning rate, 213, 217 management learning, 210 marginal learning curve model, 214 midpoint formula, 218 Northrop model, 213 operator learning, 201 S-Curve model, 212 slide property, 217 Standard-B model, 212 unit learning curve, 213 Wright model, 213 yield, see yield learning leases, 344, 346 least squares fit, 223 levelized cost of energy, 347, 446 liability, 288 life cycle definition, product, 4, scope, life-cycle cost influence diagram, 308 modeling, 239 scope, 240 lifetime buy, 370, 373 logistics definition, 249 delay time, 329 Los Alamos National Laboratory, 187 low specification limit, 55 lower confidence limit, 197 maintenance and maintainability corrective maintenance, 495 definitions, 242, 325, 332 maintenance contracts, 345 predictive maintenance, 495 scheduled, 64, 269 unscheduled, 64, 269 Manhattan Project, 187 manufacturing cost modeling, 15 marginal, 457 marketing, Markov models, 336, 350 Markov chain, 336 state transition diagram, 336 state transition probabilities, 336 state transition probability matrix, 337 material costs, 9, 24 material risk index, 374 materials-dominated products, 17 matériel, see availability materiel mean active maintenance time, 329 mean maintenance downtime, 329 mean preventative maintenance time, 329 mean supply delay, 329 mean time between failures, 260, 329 mean time between maintenance, 329 mean time between unit removals, 276 mean time to failure, 260 mean time to perform preventative maintenance, 329 mean time to repair, 65, 329, 333 mechanical throughput yield, 63 microprocessor, 144 Microsoft Xbox 360, 289 MIL-HDBK-217, 260 Mil-Specs, 357 mitigation of obsolescence cost approach, 373 modeling, Monte Carlo analysis, 183 availability example, 334 example, 198 experiment, 189 history, 187 implementation challenges, 194 Latin hypercube, 200 Index sample, 189 sample size, 189, 196 solution, 189 stratified sampling, 200 triangular distribution, 192 Moore’s Law, 143, 154 multichip modules, 48, 164 multicriteria analysis, 460 multivariate probability distributions, 191 Murphy yield model, 43 negative binomial distribution, 46 net present value, 478, 499 neural network based cost estimation, 105 newsvendor problem, 361 application, 366 critical ratio, 365 electronic parts, 366 lifetime buy, 366 no fault found, 156 non-recurring costs, 17, 24, 62, 164 non-repairable defects, 66 non-repairable items, 37 nonstationary process, 305 Norm Augustine, normal distribution, 191, 261, 278 number-up, 26, 33, 98 object points, 426 objective probability, 479, 490 object-oriented programming, 426 obsolescence, 355, 378, 429 aftermarket, 358 bridge buy, 369 budgeting/bidding support, 376 critical skills loss, 377 definition, 355 design refresh, 369 diminishing manufacturing sources and material shortages, 281, 355 549 electronic part, 357 emulation, 358 human skills, 377 inventory, 281, 355 last-time buy, see bridge buy lifetime buy, 358, 359 managing, 358 material risk index, 374 mitigation strategies, 358 organizational forgetting, 377 pro-active management, 358 reactive management, 358 return on investment, 376 risk, 375 skills obsolescence, 377 software, 377 strategic management, 359, 368 technology, 355 value, 376 Office of Management and Budget, 247 operating empty weight, 95 operation and support, 5, operational hours, 330 operator learning, 210 operator utilization, 23 opportunity cost, 316 overhead allocation, 81 costs, 9, 77 overstock cost, 361 panels, 26, 28, 142 parameters, 93 parametric, 93 parametric cost modeling, 93, 407 bounds on data, 100 cost estimating relationships, 93, 94, 407 definition, 15, 93 example, 97 limitations, 100 overfitting, 101 550 Cost Analysis of Electronic Systems scope of data, 101 service, 403–416 software, 417–432 parametric processing problems, 227 parts per million, 37 pass fraction, 127, 131 PC network, 241 PCMCIA cards, 28 performance costs, 66 performance-based logistics, 347 pick & place, 28 point defects, 227 Poisson approximation to the binomial distribution, 39, 41 distribution, 129 process, 272 yield model, 42 present value, 246 price, 8, 64 Price yield model, 45 printed circuit board, 26, 163 test, 471 printers, 433 process capability index, 54 process step, 19 calculations, 22 cost, 21 defects, 22 definition, 19 energy, 22 example, 29 fabrication/assembly, 22 inputs, 21 insertion, 22 mass, 22 material content, 22 material wasted, 22 outputs, 21 rework, 22, 160 scrap, 22 sequence, 21, 62 test/inspection, 22, 129 time, 21 waste disposition, 22 process-flow analysis, 19, 159 branch, 20 definition, 19 example, 47 examples, 27, 29, 47 test/diagnosis/rework, 160 producibility, 54 product change notice, 439 product service systems, 346, 403 production, productive time, 86 profit, prognostics and health management, 392, 495 canaries, 395 data-driven, 395 purchase price, 25 qualification, 5, 262 quality, 251 quality costs, 35 appraisal, 35 external failure, 36 internal failure, 36 prevention, 35 queuing system, 341 quote, 15 Rand Corporation, 94 random number, 191 pseudo-random numbers, 194 random sampling, 190, 203 from a data set, 193 rare event class, 466 definition, 466 importance sampling, 466 infrequent events, 465 majority-class event, 466 Index minority-class event, 466 particle splitting, 466 receiver operating characteristic, 467 unbalanced classes, 466 unbalanced misclassification costs, 466 raw coverage, 120 readiness, 348 real options analysis American options, 482 binomial lattice example, 487 binomial lattices, 485 binomial lattices – multiple time periods, 488 Black-Scholes formula, 491 correlating Black-Scholes to binomial lattice, 494 definition, 481 European options, 482 expansion option, 492 financial option, 481 futures contract, 481 in the money, 479 maintenance options, 495 management flexibility, 480, 499 path, 496 portfolio definition, 483 put option, 481 replicating portfolio theory, 483 risk-neutral probabilities, 483, 490 risk-neutral probability, 486, 490 simulation-based, 495 strike price, 479, 484 valuation, 482, 498 rebate, 299 receiver operating characteristic, 467 area under curve, 468 recurring costs, 8, 62 recycled, 155 redesign, 369 reflow, 28 rejection method, 191 551 reliability, 36, 251, 255, 257, 325 bathtub curve, 254, 307, 313 conditional reliability, 261 constant failure rate, 253 cost, 264, 460 failure distributions, 256–261 failure rate, 254, 314 FIT rates, 441, 444 hazard rate, 258 infant mortality, 253 MIL-HDBK-217, 260 unreliability, 255 useful life, 253 vs quality, 252 vs safety, 251 wear-out, 252 remaining useful life, 495 removal rate, 272 renewal function, 273, 292, 293, 327 accumulation, 319 asymptotic approximation, 296 conditioned, 305 constant failure rate, 295 delayed renewal process, 293 density function, 295, 327 functional renewal equation, 294 non-parametric renewal function estimation, 296 ordinary renewal process, 293 Weibull distribution, 297 renewals, 292 Rent’s Rule, 154 repair, 158 repairable defects, 66 replacement rate, 272 re-qualification costs, 373 requirements, 5, 6, 262 residual value, 64 return on investment, 381, 460 burn-in, 318 cost avoidance, 391 cost reduction, 383 552 Cost Analysis of Electronic Systems cost savings, 383 definition, 381 discounted cash flow, 387 failure mitigation activities, 465 flip chip example, 385–391 history, 381 manufacturing equipment replacement, 383 obsolescence, 376 stochastic, 396 technology adoption, 385 review period, 279 review time, 279 rework, 113, 155, 158 attempt, 159, 166 cost, 177 definition, 158 multi-pass example, 163 single-pass example, 160 variable rework cost and yield models, 169 risk, 247, 449, 460 continuous risk model, 462 cost, 460 cost-based FMEA, 465 definition, 460 discrete risk model, 462 mishap cost, 461 mitigation activities, 464 projected cost of failure consequences, 462 return on investment, 465 scenario-based FMEA, 465 severity level, 462 technology insertion, 460 safety, 251, 461 safety-critical systems, 345 sales, salvage, 155 sampling without replacement, 125 schedule slip, 317 scheduled maintenance, 64 scrap, 22, 122, 155 scrap fraction, 129, 131 Seeds yield model, 45, 145 SEMATECH, 61 SEMI E35, 61 sequence, 21, 62, 87 service, 403 application, 407 contract, 415 contract length, 409 example, 405 servitization, 406 should-cost, 245 Simpson distribution, 43 simulated neural network, 105 simulation, 15 SMT capacitor, 446 socket, 274, 350, 395 software, 417 adjusted function point count, 423 algorithmic models, 418 annual change traffic, 428 COCOMO, 419 cost drivers, 421 delivered source instructions, 418 development costs, 418 effort, 419, 424, 426 example, 424 feature points, 423 function point complexity weights, 424 function-point counting, 422 maintenance staffing, 428 object point analysis, 426 obsolescence, 377 productivity, 420 source lines of code, 417 support, 427 technical complexity factors, 424 technical complexity-weighting factor, 423 Index unadjusted function point count, 423 Software Productivity Research, Inc., 426 solder bumps, 386 sparing, 269 availability, 338–344 backorders, 339–341 challenges, 270 cost, 278 definition, 269, 271 economic order quantity, 279 Erlang-B, 341, 342 example, 280 inventory, 271 kit, 275 large k, 277 number of spares, 271 permanent, 273 probability of sufficiency, 274 protection level, 274, 275, 276 repairable items, 274 rotable, 274 stock-out, 271, 278 specification, spiral development process, 422 stakeholders, 243 standard error of the mean, 197 standard normal CDF, 333 standard normal statistic, 198 Stapper yield model, 45 stationary process, 305 stock-out probability, 341 stopping criteria, 197 stratified sampling, 200 sudden obsolescence, see inventory obsolescence surface mount, 28 sustain, 242 sustainability, 242 sustainment, 241 costs, 64 definition, 242 553 sustainment-dominated systems, 243, 355, 378 technology, 242 Taylor series expansion, 40 technical cost modeling, 31 technology insertion/adoption cost of risk, 461 return on investment, 385 telephone networks, 341 test, 35 automatic test equipment, 141, 149 bonepile yield, 137 built-in self test, 144 cost dependency tree, 140 defects introduced by test, 132 dependency tree, 140 design for test, 140, 143 diagnostic test, 156 economics, 113 environmental test, 113 equipment, 114, 145, 146, 164 escapes, 132 false positives, 133 fault coverage, see fault financial models, 139 functional test, 156 independent defect mechanisms, 138 integrated circuits, 113 patterns, 121 process-flow model, 129 raw coverage, 120 recurring functional, 113, 252 segments, 149 testable coverage, 120 throughput, 142 type I tester error, 133 type II tester error, 132 wafer probe, 140 test steps cascading, 138 false positives, 135 554 Cost Analysis of Electronic Systems multiple steps, 137 outgoing yield, 127 parallel, 138 test/diagnosis/rework, 113, 155, 159 example, 171 test-dominated products, 18 testers, see test equipment thermal uprating, 358 ThinPak, 174 through-life cost, see life-cycle cost throughput, 23, 62 throughput rate, 142 time value of money, 246 discount factor, 246 discount rate, 246 interest rate, 246 present value, 246 time-driven activity-based costing, 84 activity base time, 85 activity cost pool, 85 capacity cost rate, 85 duration drivers, 84 transaction drivers, 84 time-to-market, 148 tooling cost, 8, 24, 153 top-down, 19, 94, 422 total cost of ownership, 240, 433 electronic parts, 437 printers, 433 touch time, 23 trade-off analysis, traditional cost accounting, 80 traffic intensity, 341 training costs, 177 transactional drivers, 81 triangular distribution, 43, 192 truncated normal distribution, 196 unavailability, 274, 349 uncertainties aleatory, 184 data, 183 definition, 183 epistemic, 184 measurement uncertainties, 183 model uncertainty, 184 parametric, 183 service, 415 subjective uncertainties, 183 taxonomy, 183, 184 uncertainty modeling, 185 analytical methods, 185 computer algebra-based methods, 185 model resolution, 185 sampling-based methods, 185 sensitivity testing, 185 understock cost, 361 Underwriter Laboratories, 264 uniform distribution, 44, 45 unreliability, 255, 257 unscheduled maintenance, 64, 65 upper confidence limit, 197 uptime, 326 usage rate, 280 utilization, 63 value of a statistical life, 456 example, 458 hedonic valuation, 457 revealed preference methods, 457 stated preference methods, 457 variable cost, 62 variable costs, see recurring costs variate, 191 verification, volatility, 492 wafer, 26, 27, 38, 42, 98 diameter, 27 fabrication, 30 number of die on, 26, 27 probe, 140 warm up, 65 Index warranty cost models (simple), 297 definition, 287 denied warranty claims, 299 explicit, 291 first-time warranty claims, 299 fraudulent claims, 299 history, 288 implicit, 291 investment of the warranty reserve fund, 301 lifetime, 291 lump-sum rebate models, 303 non-renewing, 291 ordinary free replacement, 291 period, 291, 298, 299 pro-rata, 291, 299 renewal function, 292 renewing, 291 reserve fund, 289, 297 service costs, 307 two-dimensional, 303 types, 291 unlimited free replacement, 291 usage rate, 305 wash, 4, 393, 461 waste disposition cost, 25 waterfall process, 422 wear-out, 252 Weibull distribution, 260, 296 weighted average cost of capital, 247, 478, 523 risk-adjusted discount rate, 478 Wilson formula, see economic order quantity wind turbine, 496 wire bonding, 386 555 wirebond, 114, 116, 117 workflow modeling, 19 yield, 29, 35, 118, 127, 252 accumulation, 46, 47 composite, 46, 62 definition, 36, 38 example, 47 layered, 46 outgoing from test, 122, 128 prediction, 37 process flow example, 47, 48 relation to fault coverage, 122 yield learning, 227 defect density learning, 231 Gruber’s learning curve, 228 Hilberg’s learning curve for yield, 229 yield model exponential, 45 half Gaussian, 45 Murphy, 43 Poisson, 42, 43 Price, 45 Seeds, 45, 145 Stapper model, 45 uniform, 44 yielded cost, 50 auxiliary costs, 52 itemized, 51 ommision, 52 step yielded cost, 51 yield-loss cost, 63 z score, 277 ... Disposal Financing (cost of money) Qualification/certification Refresh/Redesign Fig 1.4 The scope of cost analysis (after [Ref 1.5]) 8 Cost Analysis of Electronic Systems 1.4 Cost Modeling Definitions... the real cost of electronic products Various manufacturing cost analysis methods are included in the book: process-flow cost modeling and parametric, cost- of- ownership, and activity-based costing... total cost of ownership of the printer For products such as aircraft, the operation and support costs can represent as much as 80% of the total cost of ownership Since manufacturing cost and the cost