References | QT Midterm

What was analysed to build this site, what each source supplied, and — just as importantly — what could not be verified.

Course documents

File	Pages	What it supplied
`CP_QT_2026-27.pdf`	10	Course code 703TO0C002, the full 7-unit syllabus and session plan pp. 6–8, the prescribed textbook list, and the assessment scheme

Prescribed text (T1): Anderson, Sweeney, Williams, Camm, Cochran, Fry & Ohlmann (2021), An Introduction to Management Science, 14th edition, CENGAGE.

Reference books: Hillier & Lieberman, Introduction to Operations Research (10th ed.); V. K. Kapoor, Operations Research, Problems & Solutions; Taha, Operations Research: An Introduction (8/e); Gupta & Hira, Operations Research. None of these books was supplied as a file, so nothing on this site is attributed to them. Every page reference here points to a document that is actually in your folder.

Notes and lecture material

File	Pages	What it supplied	Used in
`QT Complete Study Notes.pdf`	39	The backbone. Sections 0–9: what QT is, LPP building blocks and properties, step-by-step formulation, graphical method, five special cases, full simplex procedure with a three-iteration worked example, ILP types, sensitivity analysis, duality with two worked examples, and an exam quick-reference summary.	Units 1 & 2
`Linear Programming Problem.pdf`	25	Solution vocabulary p. 3, four worked formulations pp. 4–7, eight graphical problems, then slack/surplus and the simplex tableau with the pivot element pp. 19–25	Unit 1
`LPP graphical 2025.pdf`	8	Five fully worked graphical problems including Par Inc. golf bags and the diet problem, each with corner points and Z evaluated at every one	Unit 1
`Special cases in LPP.pdf`	7	Degenerate, unbounded, infeasible and multiple-optima cases, each with its identification signal and a worked example	Unit 1
`Sensitivity and Duality.pdf`	24	Reduced cost, objective coefficient range, shadow price, binding vs non-binding pp. 6–14; the advertising-spaces example with seven questions and answers pp. 15–18; duality conversion pp. 19–24	Unit 1
`Integer Programmin.pdf`	11	Why ILP exists, the three types, and seven worked formulations — toy company, project selection, shift allocation, cutting stock, multi-resource selection, fixed-charge warehouses, and logical dependencies	Unit 2

Your own class notes

These five files are handwritten or scanned, so they carry no extractable text. Every page was read as an image to work out exactly what was taught and how far each topic went.

File	Pages	What is in it
`New Note 2.pdf`	23	The main class notebook: LPP four blocks and the four formulation steps p. 2, duality pp. 11–12, assignment and transportation formulation pp. 12–18, and the three ILP types with a Mixed-ILP batching example pp. 19–21
`Linear Programming _260902_130801.pdf`	23	Graphical solutions plotted on graph paper with corner points labelled pp. 7–15; the infeasible-solution example and the start of duality pp. 19–21
`Integer Linear Programming _260901_114717.pdf`	10	Shift-allocation and cutting-stock formulations p. 2; binary and fixed-charge warehouse problems with linking constraints p. 5
`Assignment Problems_260825_081606.pdf`	6	The complete Unit 3 material: balance rule and dummy rows, 4 × 4 assignment converted to a full LPP, and a 3 × 4 transportation problem converted the same way. Pages 5–6 are blank.
`Sensitivity Analysis _260824_103438.pdf`	3	Interpretation shorthand for a solved sensitivity report — coefficient ranges, slack = 0 meaning full utilisation, dual value as economic impact. Pages 2–3 are nearly empty.

Question papers and synoptic answer keys

File	Pages	Role
Final Exam 2025-26 — Question Paper	6	26 Nov 2025, 2 hours, 50 marks. Q1 compulsory + any 4 of 6.
Final Exam 2025-26 — Synoptic Answer Key	8	Model answers and the component-by-component mark distribution — the single most useful document for learning the expected answer format.
Re-Exam 2025-26 — Question Paper	5	03 Feb 2026, same structure.
Re-Exam 2025-26 — Synoptic Answer Key	7	Model answers and mark splits for the Re-Exam.

All 14 Unit 1–3 questions from both papers are worked in full on the Solved Papers page, each with a screenshot of the original question and a check against the key.

Discrepancies found — and how they were handled

These are recorded rather than smoothed over. Each was found by working the question independently and then comparing with the printed key.

1 — Unit 3's solution algorithms are not in your notes

The session plan lists North-West Corner Rule, Least Cost Method, VAM, MODI, degeneracy, maximisation problems and the Hungarian Method for Unit 3 Policy p. 7. None of them appears anywhere in the notes. The Unit 3 material stops at formulation — identify the type, check balance, count variables and constraints, write the LPP.

Handled by: covering formulation thoroughly and stating the gap prominently on the Unit 3 page rather than inventing algorithm content you were not taught. Both papers' Unit 3 questions were pure formulation, which supports the scope.

2 — Typo in the Re-Exam synoptic key, Q1A

The key prints the first dual constraint as −3Y₁ + 2Y₂ + Y₃ − Y₄ ≤ 5 Key p. 1. The coefficient of X₁ in the primal constraint X₁ + X₂ ≥ 4 is 1, so the dual coefficient on Y₂ must be 1. The correct derivation is shown step by step.

3 — Two arithmetic slips in the Final key, Q5B

The key prints two idle-time tables that disagree with each other Key p. 5. Checked against the timetable: cell 102 → 203 should be 4.5 (key says 4.25), and cell 104 → 202 should be 15.25 (the key's two tables say 14.25 and 3.25). Full recomputation shown.

4 — Re-Exam Q4A contradicts its own table

The stem says "Projects B and D cannot be executed together" but the table lists projects P, Q, R, S, T QP p. 3 — leftover wording from the Final's version. The key's answer for this question is printed as an image with no extractable text, so the intended labels cannot be confirmed. Handled by stating the assumption explicitly, which the paper's own instruction 7 permits.

5 — No YouTube links are included

You asked for video links where they could be found. They could not be verified from this environment. Web search and navigation to video sites are not available in this session, and the technique used for the Macroeconomics site — checking each link against YouTube's oEmbed API to confirm the video exists and to read its real title and channel — could not be run here either.

Rather than print links that might be dead or point at the wrong topic, none are given. This is a gap to fill later, not a judgement that videos are unhelpful — LPP graphical method, the simplex tableau and duality conversion are all topics that benefit from watching someone work through them. Say the word when link-checking is available and they can be added and verified properly.

6 — Minor inconsistencies inside the lecture decks

The ILP deck writes the budget line of problem #7 as = 100 p. 10; a budget is a ceiling, so ≤ 100 is the defensible reading. Flagged in Unit 2, Pattern G.
The shift-allocation problem is written with ≤ 10 in the deck p. 6 and = 10 in the class notes p. 2. Both are defensible; the difference is whether every available worker must be assigned. Flagged in Pattern C.
Final Q5B's stem refers to "flight 905" while its table lists flight 105 QP p. 4. The key's own working uses both numbering schemes. It does not affect the formulation.

How this site was built

Reading. Text was extracted page by page from the seven text-bearing PDFs. The five handwritten files carry no text layer, so all 65 of their pages were rendered and read as images to establish what was actually taught and how far each topic went.
Source viewer. scripts/render_sources.py renders every page of all 16 documents to a ~1000 px JPEG — 215 pages in total. scripts/gen_manifest.py then writes js/sources.js, which the viewer reads. Re-run both after adding any new citation target.
Question screenshots. The 14 in-scope questions were located by their exact coordinates in the PDF text layer (pdftotext -bbox-layout) rather than by eye, then cropped from the rendered pages and auto-trimmed of whitespace.
Solutions. Every one was worked independently first, then compared against the synoptic key. Agreements are noted; the four disagreements above are flagged with the corrected working.