execution-plan/weeks/week-03-statistical-batch-monitoring/materials.md · UPDATED 2026-10-01

Week 03 — Materials: what to read, what to watch

6 parts892 words

Status: PREPARED 2026-09-15. Links checked on 2026-09-15 unless marked (verify). Authority: v6 master plan, Week 3 "Learn it" block (units B3 microbial growth, M3 ISLP ch. 2-3), expanded into exact locators per the source contract.

Week 03 brief · Commitment · Evidence · Review · Previous · Next

Where this fits in the week

Slot	Budget	Material
Monday build block	20 min max	R3 NIST handbook, the three pages named below. Read it while writing the T² code, not before
Tuesday · B3	45 min	R1 growth-curve section (30) → draw the curve by hand (15). The rest of the block is build
Wednesday build block	10 min	R4 `erdogant/pca` README, to sanity-check your limits against someone else's
Thursday · M3	60 min + 30	V4 eigenvectors (17) → V2 (7) → V1 (13) → one regression on two IndPenSim variables (~20); last 30 min founder paper
Saturday sandbox	optional	"Deeper", after catch-up

Hard rule for this week: limits are computed from the normal batches, never typed in. Every source below is there to defend that sentence.

↑ Back to top

Read

#	Source	Exact scope	Min	Why this week	Produce
R1	OpenStax Microbiology, ch. 9 "Microbial Growth" — 9-introduction, summary at 9-summary	The growth-curve section only (lag, log, stationary, death). Oxygen, temperature and pH are parked	30	You will discover your week-01 Monod model already explains the log phase — and what it does not explain	`notes/bio/B3.md` + a hand-drawn growth curve photographed into the notes
R3	NIST/SEMATECH e-Handbook of Statistical Methods, ch. 6	6.3.4 What are multivariate control charts? → 6.5.4.3 Hotelling's T² → 6.5.4.3.6 Constructing multivariate charts	20	The public, citable reference for the limits you compute on Monday. Phase I vs Phase II is the distinction that stops you fooling yourself	The limit formula, with its source URL, in a comment above the code that implements it
R4	erdogant/pca README	The Hotelling T² / SPE outlier section	10	A second implementation to compare against. Compare numbers, never copy them without understanding	One line in `review.md`: do your limits agree with theirs, and if not, why

↑ Back to top

Watch (YouTube)

#	Channel · title	~Length	Link	When	Produce
V1	StatQuest · "Linear Regression, Clearly Explained!!!"	~13 min	youtube.com/watch?v=7ArmBVF2dCs	Thursday, after V2	One regression fitted on two IndPenSim variables and read aloud: slope, intercept, what it means
V2	StatQuest · "Machine Learning Fundamentals: Bias and Variance"	~7 min	youtube.com/watch?v=EuBBz3bI-aA	Thursday, after V4	The bias-variance sentence in `notes/ml/M3.md`
V4	3Blue1Brown · "Eigenvectors and eigenvalues" (Essence of Linear Algebra ch. 14)	~17 min	youtube.com/watch?v=PFDu9oVAE-g (verify)	Thursday, first	One sentence in `notes/ml/M3.md`: what last Friday's principal components actually are

↑ Back to top

Deeper — Saturday sandbox only

James, Witten, Hastie, Tibshirani & Taylor, An Introduction to Statistical Learning with Applications in Python — free PDF at statlearning.com — Ch. 2 (whole) and 3.1-3.3 only. Moved out of core on 2026-09-29.
Stanford Online · Statistical Learning with Python (Hastie, Tibshirani et al.) — Playlist: youtube.com/playlist?list=PLoROMvodv4rPP6braWoRt5UCXYZ71GZIQ · course page: online.stanford.edu/courses/sohs-ystatslearningp-statistical-learning-python. Moved out of core on 2026-09-29.
Nomikos P. & MacGregor J.F., "Multivariate SPC Charts for Monitoring Batch Processes", Technometrics 37(1):41-59, 1995. DOI 10.1080/00401706.1995.10485888 · author PDF mirror: three-mode.leidenuniv.nl/pdf/n/nomikosmacgregor1995tmet.pdf. The founding paper of batch MSPC — abstract and figures are enough. Cite it in week 08's report.
Kourti T. & MacGregor J.F., "Process analysis, monitoring and diagnosis, using multivariate projection methods", Chemometrics and Intelligent Laboratory Systems 28:3-21, 1995. (verify)
salvadorgarciamunoz/pyphi — a pharma practitioner's PCA/PLS with T² and SPE. Niche, but written by someone who does this for a living.
Coursera Introduction to Statistics (Stanford, Walther), regression module — coursera.org/learn/stanford-statistics, free audit. Only if ISLP ch. 3 feels steep.

↑ Back to top

Link check log — 2026-09-15

Verified today: R1 (both pages), R3 (all three NIST pages), V1, V2, V3 playlist and Stanford course page, Nomikos & MacGregor DOI and PDF mirror.
Not re-checked today: R2 statlearning.com (verified 2026-09-12, see SOURCES.md), R4, pyphi, Kourti & MacGregor (verify), Coursera statistics.
Record in evidence.md which NIST page each computed limit came from. A limit without a citation is a threshold someone typed.

↑ Back to top

NotebookLM kit sources for `LABLOOP-W03`

(1) this file; (2) the Week 3 chapter from the v6 master; (3) R1 pages; (4) ISLP ch. 2-3 PDF; (5) V3 ch. 2 and ch. 3 video URLs; (6) the three NIST pages; (7) V1 and V2 URLs; (8) Nomikos & MacGregor abstract; (9) your T² plot description and computed limits as text; (10) QUESTIONS.md.

Focus line: Focus: from a golden batch anecdote to a statistical envelope; Hotelling T² and SPE as two different ways a batch can be abnormal; the bias-variance trade-off.

Custom question: Explain the difference between T² and SPE alarms using a portfolio-dashboard analogy, then tell me which of the two should fire first for a feed-pump failure and why.

↑ Back to top