import type { CpkInput, CpkResult } from './types.js';
/**
 * Calculate Cpk (Process Capability Index)
 *
 * @formula
 *   - Cp  = (USL - LSL) / (6σ)           — Process potential
 *   - Cpu = (USL - mean) / (3σ)          — Upper capability
 *   - Cpl = (mean - LSL) / (3σ)          — Lower capability
 *   - Cpk = min(Cpu, Cpl)                — Actual capability
 *   - Sigma Level = 3 × Cpk
 *   - Within-Spec % = [Φ((USL−mean)/σ) − Φ((LSL−mean)/σ)] × 100
 *
 * @reference Montgomery, D. C. "Introduction to Statistical Quality Control", 7th Ed. Wiley.
 * @reference ISO 22514-2:2017. Statistical methods in process management — Capability and performance.
 *
 * @units USL/LSL/mean/stdDev: same unit (e.g., mm, g); Cp/Cpk dimensionless; withinSpecPercent in %
 *
 * @remarks
 *   `withinSpecPercent` estimates the proportion of output within [LSL, USL] assuming a normal
 *   distribution with the supplied (short-term) σ. It mirrors `ppk()`'s field so the
 *   capability (`cpk`) and performance (`ppk`) APIs stay symmetric; note that for Cpk this is a
 *   short-term/potential estimate, whereas `ppk()` uses long-term σ.
 *
 * @validation
 *   - Cpk ≥ 1.33: Capable process (4σ from nearest spec)
 *   - Cpk ≥ 2.00: Six Sigma process (6σ from nearest spec)
 *   - Cpk = Cp when process is perfectly centered
 *   - stdDev ≤ 0: returns a zero-valued result (no meaningful capability); does not throw
 *
 * @param input - Cpk input parameters
 * @returns Cpk result with capability indices, sigma level, and within-spec percentage
 */
export declare function cpk(input: CpkInput): CpkResult;
//# sourceMappingURL=cpk.d.ts.map