23-Chem-A5 Chemical Plant Design and Economics · May 2013
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
National Exams — May 2013 — 04-Chem-A5 Chemical Plant Design and Economics. Three-hour, open-book exam; any non-communicating calculator permitted. The paper poses seven equally weighted essay questions and the candidate answers any five; only five are marked. All seven are answered below for completeness. These are conceptual design-and-economics questions — the solutions are written as organised prose (clarity and organisation are explicitly marked). The one numerical illustration (a Canadian Capital Cost Allowance schedule in Q2) is worked from stated assumptions.
Reference texts: M.S. Peters, K.D. Timmerhaus & R.E. West, Plant Design and Economics for Chemical Engineers (5th ed., McGraw-Hill) — the exam's named primary text (cost estimation, profitability, depreciation, optimisation); R. Turton et al., Analysis, Synthesis, and Design of Chemical Processes (4th ed., Prentice Hall) — process synthesis, safety, and economics; W.D. Seider et al., Product and Process Design Principles (3rd ed., Wiley) — separation-train synthesis and heuristics; supporting Canadian tax practice from the Canada Revenue Agency Capital Cost Allowance classes and the half-year rule.
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
A profitability estimate (NPV, discounted cash-flow rate of return, payback) is only as reliable as the forecasts fed into it — product price, feedstock cost, plant capacity, capital cost, and start-up date — and over a five-year lead time every one of those can move. Formal risk-assessment procedures replace a single deterministic answer with an understanding of how the answer behaves when the inputs are uncertain. Two widely used procedures are described below.
Procedure 1 — Sensitivity analysis. Here each uncertain variable is changed by a chosen amount (say $\pm 20\%$) one at a time while the others are held at their base values, and the resulting change in the profitability measure (NPV or DCFROR) is recorded. Plotting all variables together produces a "spider" or "tornado" diagram whose steepest lines identify the factors to which profitability is most sensitive — commonly product selling price and raw-material cost. The value of the method is diagnostic: it tells management where to concentrate effort (tighten a price contract, lock in a feedstock supply, refine a capital estimate) and how much margin of error the project can absorb before the return falls below the acceptable minimum. Its limitation is that it varies factors singly and says nothing about how likely a given deviation is.
Procedure 2 — Probabilistic (Monte-Carlo) risk analysis. This procedure addresses that limitation by assigning a probability distribution to each uncertain input — for example a triangular distribution on capital cost with pessimistic, most-likely and optimistic values, and a distribution on price reflecting market volatility. A computer then draws thousands of random samples from these distributions, evaluating the cash-flow model for each combination. The output is a full distribution of NPV rather than a single number, from which one reads the expected value, the spread, and directly the probability that the project loses money (the fraction of trials with NPV < 0). This converts "risk" from a qualitative worry into a quantified probability that can be compared across competing projects and set against the corporation's risk tolerance.
Both procedures are routinely supplemented by scenario analysis (coherent best-/most-likely-/worst-case bundles), break-even analysis, and, for staged commitments, decision-tree / expected-value methods — but sensitivity analysis and Monte-Carlo simulation are the two workhorses of project economic risk assessment.
