Skip to content
All models

Declared uniform distribution model

Seeded tolerance sampling

Inputs

Nominal one-dimensional result before sampled contributor variation.
User-declared symmetric uniform half-width; enter 0 only when unused.
User-declared symmetric uniform half-width; enter 0 only when unused.
User-declared symmetric uniform half-width; enter 0 only when unused.
Optional visible contributor; enter 0 when unused.
Optional visible contributor; enter 0 when unused.
Optional visible contributor; enter 0 when unused.
Visible non-negative integer state for reproducible local pseudo-random draws; it does not represent a physical population.
Finite number of local draws from 10 through 10,000; more draws do not validate the input model.

Results

Generated local samples

1,000

samples
Reproducibility seed

20,260,819

Generated sample mean

99.998

Generated sample standard deviation

0.065179

Generated sample minimum

99.828

Generated sample maximum

100.16

Generated 1st percentile

99.864

Generated 99th percentile

100.13

Declared worst-case envelope half-width

0.18

xi = N + Σ[(2ui−1)·ti], ui from visible seeded LCG · reported percentiles use nearest generated rank

A first-pass number, not a code check, certification, or approval. Read the method and its limits below. What this is and is not.

Nearby: ISO 286 fits · Tolerance stack · Position deviation · MMC bonus & virtual condition

Method

Formula, when it applies, and when it does not

Generate a reproducible local sample of a six-contributor linear stack using symmetric uniform half-widths, a visible seed, and a visible sample count.

xi = N + Σ[(2ui−1)·ti], ui from visible seeded LCG · reported percentiles use nearest generated rank

When

  • Linear one-dimensional additive stack
  • Uniform distribution selected by the user
  • Independent seeded pseudo-random draws
  • Visible integer seed and finite sample count

Don’t

  • This is a reproducible local pseudo-random sample of a user-declared independent uniform linear stack. It does not infer contributor distributions or independence, model geometry or assembly, estimate a physical population, predict yield, determine capability/compliance, recommend tolerances, or approve an assembly.
Root-sum-square statistical tolerance stacking — standard result