Library / Engineering / KEVOS® original book

First-of-a-Kind Technology by the Numbers

Readiness levels, pioneer plants, scale-up, qualification, ramp-up and novelty-graded contingency, worked through one recycler's choice

Measure what novelty costs: readiness levels, pioneer-plant evidence, scale-up, qualification tests, ramp-up curves, contingency graded by novelty, and ways to buy the risk down.

10 chapters · About 101 minutes · Free to read · Intermediate process engineering, project and investment practice

Start reading
KEVOS® LibraryENGINEERINGFirst-of-a-KindTechnology bythe NumbersKEVOS® ORIGINAL BOOK10 CHAPTERS

About this book

A process that has never run at full scale is a promise backed by a pilot. It may well be better than the equipment it would replace, and its makers usually have good evidence that it works. What they cannot yet have is evidence that it works at ten times the size, for thousands of hours, on someone else’s feed, with someone else’s operators. The first full-scale plant of a technology is where that evidence is produced, and the owner of the first plant pays for it: in capital cost that grows during the project, in a longer schedule, and above all in a start-up that takes months or years to reach the output the business case assumed. A plant built to a design that has run many times before carries much less of that risk, because others have already paid for the learning.

This book works through that difference with numbers. It starts with first-of-a-kind and nth-of-a-kind plants and what the published evidence says about them, then rates technology readiness element by element, explains from research on pioneer process plants where cost growth and performance shortfalls come from, and places suppliers on their experience curves. It goes on to scale-up from pilot to production, qualification tests sized to show reliability and yield on the owner’s own feed, ramp-up and reliability growth curves, and cost and schedule allowances that grow with novelty. It ends with the ways of buying the risk down, qualifying first, guarantees and separable modules, valued against each other, and with the choice and what to watch after it. Each chapter follows one running case, includes a plausible mistake and how it was found, and ends with an exercise and its worked answer.

It is for owners, managers, process engineers and finance staff who must choose between proven equipment and a new process, or approve the first full-scale plant of a technology. It assumes spreadsheet arithmetic, logarithms and net present value at the level of a business case; readiness scales, reliability bounds and simulation are introduced where they are needed.

The running case

Every chapter follows one fictional decision. This is an illustration. A family-owned plastics recycler in Perth’s southern industrial area, with about 55 staff, collects, sorts and bales post-consumer rigid HDPE and PP packaging and sells the bales to processors. In March 2024 its board must choose a washing-and-sorting line to turn 18,000 t a year of bales into washed, sorted flake. Option A is a proven line from an established supplier that has built 41 similar lines, guaranteed at 72% yield, with a capital cost of A$16.0 million. Option B is a novel process from a small Australian technology company, claiming 84% yield from a hydrocyclone wash, a closed-loop water system and a hyperspectral fines sorter, for A$17.5 million, and proven so far on a pilot one-tenth of the size.

On paper the novel process is far better: about A$2.4 million a year more contribution for A$1.5 million more capital. The chapters test that claim one element at a time, through readiness, scale-up, qualification, ramp-up and contingency, and follow the board to its decision and the first months of operation.

The business, its suppliers, its plant, its people and every number in the case are invented. None describes KEVOS® work, a real business, a real supplier or a real technology. All amounts are Australian dollars excluding GST.

How to use the book

Read Chapter 1 first, because it sets out the two options and the paper comparison that the rest of the book corrects. Chapters 2 to 4 measure novelty and experience, Chapters 5 to 7 follow the technology from pilot to steady operation, and Chapters 8 to 10 turn the findings into allowances, risk reduction and a decision. The corrected yield found in Chapter 6 is used in every later chapter. Every calculation was checked by running it, and the KEVOS® tools linked in the chapters reproduce the learning-curve, contingency and schedule arithmetic.

The book is one of a series of KEVOS® Library books on how projects are planned and delivered. Reference Class Forecasting by the Numbers supplies the outside view and the percentile and simulation methods that Chapter 8 relies on. Front-End Planning by the Numbers covers project definition and estimate classes, one of the drivers of cost growth in Chapter 3. Stage Gates and Stopping Rules by the Numbers develops the value of information and real options behind Chapter 9, and Project Teams and Contracts by the Numbers covers the contracts through which guarantees and risk are allocated.

Source limits and professional responsibility

The methods are established and the sources are listed with what each was used for. Where the book quotes published results, they are attributed and concern the plants and programmes that research studied, mostly large North American process plants and US defence programmes; their relevance to a small Australian plant is discussed where they are used. The readiness ratings, cost-growth factors, ramp-up curves and probabilities in the case are the book’s own teaching values, not published rates, and a real decision needs its own evidence, quotes and independent review. Recycling plant is hazardous plant, and its design, installation and operation are governed by work health and safety laws, environmental and trade-waste approvals, and the competent people responsible for them. This book is general information about technology and investment decisions. It is not engineering, environmental, legal or financial advice for any real plant.

Who this book is for

Owners, operations, engineering and project managers, process engineers and finance staff in Australian businesses who must choose between proven equipment and a new process, or approve the first full-scale plant of a technology.

What you will learn

  • Explain why first-of-a-kind plants cost more, take longer and start up more slowly than nth-of-a-kind plants, using the published evidence.
  • Rate the readiness of each critical technology element, and set system readiness from the least mature element rather than an average.
  • Check a pilot against the full-scale design flow by flow, and size a qualification test for reliability and yield on the owner's own feed.
  • Model production ramp-up and reliability growth as curves, and grade cost and schedule allowances by readiness.
  • Value qualification, performance guarantees and separable modules as ways of buying down technology risk, and set pass criteria before the trial.

Inside this book

Chapters

  1. 01The price of being first: a proven line against a novel processA recycler's proven A$16.0m washing line against a novel A$17.5m process: paybacks of 3.39 and 2.45 years, a A$13.8m paper NPV lead, and why that lead is still only a claim.
  2. 02Technology readiness levels and critical technology elementsThe nine readiness levels in plain words, critical technology elements, Option B's ratings of 6, 5 and 7, and why system readiness is TRL 5, not the 7.85 average.
  3. 03Pioneer plants: where cost growth and performance shortfalls come fromWhat RAND's pioneer-plant studies found drives cost growth and start-up shortfalls, applied to a 40% novelty fraction, open effluent scope, a recycle loop and wet fines.
  4. 04Experience curves, supplier track records and the operating teamA builder's 41st line against a first unit on a 90% curve, the novel package on an 85% curve from A$9.33m, who pays for learning, and a sorter scored on borrowed experience.
  5. 05Scale-up from pilot to production: what changes with sizeFrom a 0.30 t/h pilot to a 3.0 t/h plant: numbering-up ten cyclones, sorter loading up 2.5 times, and a water loop whose pilot never saw the plant's 3,000 mg/L.
  6. 06Qualification testing: run length, feed envelope and yield basisSize a qualification run with chi-square bounds (1,500 h for MTBF 500 h at 80%), test across a 5–13% contamination envelope, and correct a claimed 84% yield to 78.9%.
  7. 07Ramp-up: from first flake to nameplate outputExponential ramp-up curves for a proven and a novel line (2.8 against 21.9 months to 90%), Duane reliability growth, and a linear ramp that overstated contribution by A$3.46 million.
  8. 08Contingency and schedule allowances that grow with noveltyCost and schedule allowances graded by readiness: element P50s and P80s, a 10,000-run simulation giving B a P80 of A$22.9m, and the errors of a flat 10% and of adding P80s.
  9. 09Buying down technology risk: qualification, guarantees and separable modulesQualifying first (expected NPV A$11.35m), a yield guarantee whose damages cover 6.8% of the shortfall, and a separable fines module worth A$856,924 as an option, valued against doing nothing.
  10. 10Choosing between proven and novel, and following the choice throughThe risk-adjusted choice of the proven line now and a fines module later, ramp-up tracked against its curve, and a module pass criterion that a two-week clean-bale trial nearly bypassed.

Apply This in Practice

Use these KEVOS® tools to try the method from this reading with your own numbers.

References and source notes

The manuscript and worked examples are original KEVOS® writing. References inform particular concepts; their authors do not endorse this book. Web references were consulted on 28 October 2026.

  • Bent Flyvbjerg and Dan Gardner. How Big Things Get Done (2023). Supplied source, © Connaught Street Inc. and Bent Flyvbjerg, in copyright. Used as a subject map for experience, proven technology and first-of-a-kind risk only. No passages, chapter structure, stories, heuristics or figures are reproduced.
  • John C. Mankins, NASA Office of Space Access and Technology, Advanced Concepts Office. Technology Readiness Levels: A White Paper (1995). Source of the nine-level technology readiness scale with a definition for each level. Chapter 2 paraphrases the levels in its own words and notes that organisations word them differently.
  • John C. Mankins, Acta Astronautica 65(9–10), 1216–1223. Technology readiness assessments: A retrospective (2009). Background on the origin of technology readiness levels at NASA in the 1970s and their later adoption by the US Department of Defense and the Government Accountability Office, used in Chapter 2.
  • US Government Accountability Office. Technology Readiness Assessment Guide: Best Practices for Evaluating the Readiness of Technology for Use in Acquisition Programs and Projects (GAO-20-48G) (2020). Source for assessing readiness as a systematic, evidence-based process applied to the critical technologies of a larger system, used for the assessment method in Chapters 2 and 6. Its level definitions are not reproduced.
  • US General Accounting Office. Best Practices: Better Management of Technology Development Can Improve Weapon System Outcomes (NSIAD-99-162) (1999). Source for the recommendation that key technologies reach a maturity analogous to technology readiness level 7 before a programme commits to a cost, schedule and performance baseline, cited in Chapters 2 and 9.
  • US Government Accountability Office. Defense Acquisitions: Assessments of Selected Major Weapon Programs (GAO-06-391) (2006). Source for the average research and development cost growth of 4.8 per cent for programmes that began development with mature technologies against 34.9 per cent for those that began with immature technologies, quoted with attribution in Chapters 1 and 8 as evidence of direction and size, not as a rate for other industries.
  • Edward W. Merrow, Kenneth E. Phillips and Christopher W. Myers, RAND Corporation. Understanding Cost Growth and Performance Shortfalls in Pioneer Process Plants (R-2569-DOE) (1981). Study of 44 process plants built by 34 oil, chemical and mineral companies. Used in Chapters 3 and 7 for its findings that cost-estimation error is explained mainly by departure from prior technology, project and site definition, and complexity, and that performance shortfalls are explained mainly by new technology and solids processing. No coefficients are reproduced.
  • Edward W. Merrow, Stephen W. Chapel and Christopher Worthing, RAND Corporation. A Review of Cost Estimation in New Technologies: Implications for Energy Process Plants (R-2481-DOE) (1979). Background on the factors common to estimation errors across the systems reviewed: definition at the time of the estimate, scope changes and the level of technological innovation, used in Chapter 3.
  • Edward W. Merrow, RAND Corporation. A Quantitative Assessment of R&D Requirements for Solids Processing Technology (R-3216) (1986). Source for the finding that the main performance problems of solids-processing plants are physical and mechanical rather than chemical, and did not lessen in newer plants. Used qualitatively in Chapters 3, 5 and 7.
  • T. P. Wright, Journal of the Aeronautical Sciences 3(4), 122–128. Factors affecting the cost of airplanes (1936). Origin of the learning curve, used in Chapter 4 as the basis of the experience curve for whole plants.
  • Bruce D. Henderson, Journal of Business Strategy 4(3), 3–9. The application and misapplication of the experience curve (1984). Background on the experience curve and on how it is misapplied, used in Chapter 4.
  • Linda Argote and Dennis Epple, Science 247, 920–924. Learning curves in manufacturing (1990). Background on learning curves in organisations, used in Chapters 4 and 7. No results are quoted.
  • Roger E. Bohn and Christian Terwiesch, Journal of Operations Management. The economics of yield-driven processes (1999). Background on why yield matters most when capacity is constrained, as during ramp-up, used in Chapters 1 and 7. No figures are quoted.
  • Christian Terwiesch and Roger E. Bohn, International Journal of Production Economics 70(1), 1–19. Learning and process improvement during production ramp-up (2001). Background on ramp-up as a period of learning in which experiments compete with output, and early learning is worth more than late, used in Chapter 7.
  • J. T. Duane, IEEE Transactions on Aerospace 2(2), 563–566. Learning curve approach to reliability monitoring (1964). Origin of the reliability growth model in which cumulative mean time between failures plots as a straight line against cumulative operating time on log-log axes, used in Chapter 7.
  • Patrick D. T. O'Connor and Andre Kleyner, Wiley, 5th edition. Practical Reliability Engineering (2012). Reference for reliability demonstration and reliability growth methods, including chi-square bounds on mean time between failures for time-terminated tests, used in Chapters 6 and 7.
  • Australian Renewable Energy Agency (ARENA). Technology and commercial readiness tools (current online guidance). Australian example of technology readiness levels used by a funding agency, with ARENA's Commercial Readiness Index (1 to 6, from a hypothetical commercial proposition to a bankable asset class) to be read alongside them. Cited in Chapter 2.
  • AACE International. AACE International Recommended Practice 18R-97: Cost Estimate Classification System, as applied in engineering, procurement and construction for the process industries (current revision). Named by number and scope in Chapter 3 as the classification linking estimate class to the maturity of project definition. No ranges or values are quoted.
  • Commonwealth of Australia. Recycling and Waste Reduction Act 2020 and the waste plastic export rules made under it (2020 (Act); plastics requirements from 1 July 2021 and 1 July 2022). Named as context for the running case: exports of mixed waste plastics, and later of unprocessed single-resin waste plastics, were stopped unless the plastics met processing requirements. Readers should check the current requirements with the responsible Australian Government department.
  • Safe Work Australia. Model Code of Practice: Managing the risks of plant in the workplace (current edition). Named as the national model guidance for people who design, install, commission and use plant such as shredders, granulators and conveyors. Check with the local regulator whether it has legal effect in a given state or territory.

Continue exploring

← All Library books