Study guide
11.2 Statistical Process Control for Attributes (p, np, c, and u Charts)
PE Industrial and Systems Exam
Track your progress as you read
Finished sections get a check mark.
1 Exam Overview, Strategy & Systems Foundations
1.1 PE Industrial Exam Format, Official Specification & Reference Handbook Strategy 1.2 System Analysis Tools, Process Definition, Requirements Engineering & QFD 1.3 Performance Measures: Output, Productivity, OEE, Cost & Environmental Sustainability 1.4 Process Types (Discrete, Continuous, Manufacturing, Service) & Value Analysis/Engineering
2 Optimization, Queueing, Simulation, Networks & Markov Chains
2.1 Linear Programming Formulation, Graphical Solutions & Sensitivity Analysis 2.2 Queueing Theory Models (M/M/1, M/M/s) & Little's Law Applications 2.3 Discrete-Event Simulation, Random Variates & Model Validation 2.4 Graph Theory & Network Models: Shortest Path, Minimal Spanning Tree & Maximum Flow 2.5 Markov Chains: Transition Matrices, Steady-State Probabilities & Absorbing States 2.6 Decision Trees, Expected Value & Bottleneck Analysis (Theory of Constraints)
3 Project Management & Planning
3.1 Project Scheduling with CPM and PERT: Critical Path, Float, Probability & Crashing 3.2 Project Planning & Control: Scope and WBS, Gantt Charts, Critical Chain (CCPM), Risk & Earned Value
4 Facility Layout, Space, Capacity & Material Handling
4.1 Systematic Layout Planning (SLP) & Activity Relationship (REL) Analysis 4.2 Workstation Design, Aisle Sizing & Space Requirements Planning 4.3 Capacity Analysis: Machine Requirements, Scrap Allowances & Personnel Requirements 4.4 Material Handling Equipment Selection, Unit Loads & Handling Principles
5 Site Location, Warehousing & Facility Economics
5.1 Facility Location Models: Center of Gravity & Minisum / Minimax Distance 5.2 Warehousing Systems, Storage Assignment & Order Picking Operations 5.3 Facility Life-Cycle Economics, Replacement Analysis & Rearrangement Costs
6 Demand Forecasting, Production Planning, MRP & JIT
6.1 Demand Forecasting Methods: Moving Averages, Exponential Smoothing & Error Metrics 6.2 Aggregate Planning Strategies & Master Production Scheduling (MPS) 6.3 Material Requirements Planning (MRP), Bill of Materials & Capacity Planning 6.4 Just-in-Time & Lean Production Planning: Pull Systems, Kanban Sizing, Heijunka & SMED
7 Inventory Systems, Scheduling, Logistics & Transportation Modes
7.1 Deterministic Inventory Control: Classical EOQ, EPQ & Quantity Discounts 7.2 Stochastic Inventory Management: Safety Stock, Continuous / Periodic Review & Newsvendor 7.3 Production Sequencing, Dispatching Rules & Johnson's Algorithm 7.4 Supply Chain Logistics, Transportation Simplex & Distribution Network Design 7.5 Transportation Modes, Freight Economics & the Transshipment Model
8 Engineering Economics, Break-Even Analysis & Costing Systems
8.1 Equivalence, Cash Flow Analysis & Discounted Cash Flow Metrics (PW, FW, EAC) 8.2 Rate of Return Analysis (IRR, Incremental IRR) & Benefit-Cost (B/C) Ratios 8.3 Asset Depreciation (MACRS, Straight-Line), Tax Effects & Replacement Decisions 8.4 Break-Even Analysis, Technical Capability Assessment & Costing Systems (Traditional, ABC & Standard Costs)
9 Methods Engineering, Work Measurement, Learning Curves, Line Balancing & Process Planning
9.1 Methods Analysis, Process Charting & Motion Study (Therbligs, 5S) 9.2 Worker–Machine (Man–Machine) Charts, Multiple Activity Analysis & Synchronous Servicing 9.3 Time Study Fundamentals, Performance Rating & Allowance Factors 9.4 Predetermined Motion Time Systems (PMTS, MTM, MOST) & Work Sampling 9.5 Learning Curves: Unit and Cumulative-Average Models, Learning Rates & Applications 9.6 Assembly Line Balancing, Precedence Constraints & Efficiency Heuristics 9.7 Manufacturing & Service Process Planning: Operation Selection, Sequencing, Route Sheets, Fixturing & Mistake Proofing
10 Ergonomics, Human Capacity, Occupational Safety & DART Rates
10.1 Anthropometry, Percentile Design, Workstation Ergonomics & Link Analysis 10.2 Industrial Biomechanics, the NIOSH Lifting Equation & Lifting Aids 10.3 Limits of Human Capacity & Risk-Factor Tools: Work Physiology, Strength, Range of Motion, Vision, RULA & Slip Resistance 10.4 Environmental Ergonomics (Noise, Heat, Illumination), Human-Machine Systems & OSHA Safety 10.5 Safety Codes, Standards & Voluntary Guidelines (OSHA, ANSI, NIOSH, MIL-STD) and DART Rate Calculations
11 Statistical Process Control & Process Capability
11.1 Statistical Process Control for Variables (X-bar/R and X-bar/s Charts) 11.2 Statistical Process Control for Attributes (p, np, c, and u Charts) 11.3 Process Capability Indices (Cp, Cpk, Cpm, Pp, Ppk) & Sigma Levels
12 Design of Experiments, Acceptance Sampling & Quality Systems
12.1 Design of Experiments: 2^k Factorials, ANOVA & Taguchi Robust Design 12.2 Acceptance Sampling: OC Curves, AQL, LTPD, AOQL, Z1.4/MIL-STD-105E & Dodge-Romig 12.3 Quality Systems (Deming, TQM, ISO 9000, Benchmarking), Root Cause Analysis & DMAIC
13 Reliability Engineering & Maintenance Systems
13.1 Reliability Modeling, Failure Rates (lambda) & Life Distributions (Exponential, Weibull) 13.2 System Reliability Architecture: Series, Parallel, k-out-of-n & Redundancy 13.3 Equipment Maintenance Engineering, Availability (A), MTBF/MTTR & RCM
11.2 Statistical Process Control for Attributes (p, np, c, and u Charts)
Key Takeaways Explain This Topic AI
Attribute data is bifurcated into nonconforming units (defective items governed by the binomial distribution) and nonconformities (defects per unit governed by the Poisson distribution).
The p p p -chart monitors the fraction nonconforming and accommodates either constant or variable sample sizes ( n i n_i n i ), whereas the n p np n p -chart tracks the absolute count of defective units and strictly requires a constant sample size across all subgroups.
The c c c -chart tracks total defect counts over a strictly constant area of opportunity ( n = 1 n=1 n = 1 ), whereas the u u u -chart monitors defect density per unit across variable inspection units ( n i n_i n i ).
Because proportions, counts of defectives, and defect counts cannot be negative, calculated lower control limits that evaluate below zero must be truncated to zero ( L C L = 0 LCL = 0 L C L = 0 ).
Normal approximation to the binomial and Poisson distributions requires adequate expected counts ( n p ≥ 5 np \ge 5 n p ≥ 5 or c ˉ ≥ 5 \bar{c} \ge 5 c ˉ ≥ 5 ); small sample sizes yield severe skewness and asymmetric false alarm rates.
Last updated: October 2026
Statistical Process Control for Attributes ( p p p , n p np n p , c c c , and u u u Charts)
Core Principle: Attribute control charts monitor qualitative, count-based, or pass/fail quality characteristics. Attribute SPC is organized according to whether the inspection records nonconforming units (defective products modeled by the binomial distribution) or nonconformities (defects per inspection unit modeled by the Poisson distribution).
In many manufacturing, assembly, and service systems, measuring continuous numerical dimensions is neither practical nor cost-effective. Instead, quality inspectors evaluate parts using "go/no-go" functional gauges, visual optical inspection, or defect counts per assembly. Statistical Process Control for attributes provides the quantitative framework for tracking these discrete metrics. Industrial engineers must select the correct attribute chart based on the mathematical nature of the data and whether the sample inspection size remains constant or varies.
1. Fundamentals of Attribute Data: Defectives vs. Defects
A critical distinction on the PE exam is between nonconforming units (defective items) and nonconformities (defects): Attribute Data Classification
- ├── Nonconforming Units (Defective Items)
- │ ├── Binary outcome: Conforming (Pass) vs Nonconforming (Fail)
- │ ├── Underlying distribution: Binomial Distribution B(n, p)
│ └── Appropriate Control Charts:
- │ ├── p-chart: Fraction nonconforming (constant or variable n)
- │ └── np-chart: Number nonconforming (strictly constant n)
- └── Nonconformities (Defects per Unit)
- ├── Count of discrete imperfections, flaws, or blemishes per unit
- ├── Underlying distribution: Poisson Distribution Poisson(lambda)
└── Appropriate Control Charts:
- ├── c-chart: Total defects across constant area of opportunity (n = 1)
- └── u-chart: Defects per inspection unit (constant or variable n_i)
- The Binomial Model for Nonconforming Units
When an inspection station evaluates whether a manufactured part meets functional standards, each part represents a Bernoulli trial with two mutually exclusive outcomes: conforming (probability 1 − p 1 - p 1 − p ) or nonconforming (probability p p p ). In a random sample of n n n independent units, the probability of observing exactly d d d nonconforming units is governed by the Binomial Distribution :
P ( D = d ) = ( n d ) p d ( 1 − p ) n − d P(D = d) = \binom{n}{d} p^d (1 - p)^{n - d} P ( D = d ) = ( d n ) p d ( 1 − p ) n − d Expected Count: E ( D ) = n p Variance: Var ( D ) = n p ( 1 − p ) \text{Expected Count: } E(D) = np \qquad \text{Variance: } \operatorname{Var}(D) = np(1 - p) Expected Count: E ( D ) = n p Variance: Var ( D ) = n p ( 1 − p )
For the fraction nonconforming p ^ = D / n \hat{p} = D / n p ^ = D / n :
E ( p ^ ) = p Var ( p ^ ) = Var ( D ) n 2 = p ( 1 − p ) n E(\hat{p}) = p \qquad \operatorname{Var}(\hat{p}) = \frac{\operatorname{Var}(D)}{n^2} = \frac{p(1 - p)}{n} E ( p ^ ) = p Var ( p ^ ) = n 2 Var ( D ) = n p ( 1 − p )
The Poisson Model for Nonconformities
A single complex product (e.g., an automobile door panel, a printed circuit board, or an aircraft turbine blade) can satisfy overall functional requirements while exhibiting several discrete blemishes (scratches, pinholes, solder bridges). When the opportunities for defects across a continuous area of opportunity are vast, but the probability of a defect at any specific point is extremely small, defect occurrences follow the Poisson Distribution :
P ( C = c ) = e − λ λ c c ! for c = 0 , 1 , 2 , … P(C = c) = \frac{e^{-\lambda} \lambda^c}{c!} \quad \text{for } c = 0, 1, 2, \dots P ( C = c ) = c ! e − λ λ c for c = 0 , 1 , 2 , … Expected Count: E ( C ) = λ Variance: Var ( C ) = λ \text{Expected Count: } E(C) = \lambda \qquad \text{Variance: } \operatorname{Var}(C) = \lambda Expected Count: E ( C ) = λ Variance: Var ( C ) = λ
The defining characteristic of the Poisson distribution is that its variance equals its mean ( σ c 2 = μ c = λ \sigma_c^2 = \mu_c = \lambda σ c 2 = μ c = λ ).
- 2. Fraction Nonconforming ( p p p ) and Number Nonconforming ( n p np n p ) Charts
- The p p p -Chart (Fraction Nonconforming)
The p p p -chart monitors the proportion of nonconforming items in a subgroup. It is the most versatile attribute chart because it accommodates both constant sample sizes and variable sample sizes ( n i n_i n i ).
Given m m m subgroups where subgroup i i i contains n i n_i n i inspected units and d i d_i d i nonconforming units, the subgroup fraction nonconforming is p i = d i / n i p_i = d_i / n_i p i = d i / n i . The overall weighted process fraction nonconforming is:
p ˉ = ∑ i = 1 m d i ∑ i = 1 m n i \bar{p} = \frac{\sum_{i=1}^m d_i}{\sum_{i=1}^m n_i} p ˉ = ∑ i = 1 m n i ∑ i = 1 m d i
Case A: Constant Sample Size ( n i = n n_i = n n i = n )
When every subgroup contains exactly n n n units, the control limits remain fixed across all intervals:
C L p = p ˉ CL_p = \bar{p} C L p = p ˉ U C L p = p ˉ + 3 p ˉ ( 1 − p ˉ ) n UCL_p = \bar{p} + 3 \sqrt{\frac{\bar{p}(1 - \bar{p})}{n}} U C L p = p ˉ + 3 n p ˉ ( 1 − p ˉ ) L C L p = max ( 0 , p ˉ − 3 p ˉ ( 1 − p ˉ ) n ) LCL_p = \max\left(0, \bar{p} - 3 \sqrt{\frac{\bar{p}(1 - \bar{p})}{n}}\right) L C L p = max ( 0 , p ˉ − 3 n p ˉ ( 1 − p ˉ ) )
Case B: Variable Sample Size ( n i n_i n i varies)
In high-mix manufacturing or shipping dock receiving inspection, the number of inspected units varies per lot. Control limits must be calculated individually for each subgroup i i i :
U C L p , i = p ˉ + 3 p ˉ ( 1 − p ˉ ) n i UCL_{p,i} = \bar{p} + 3 \sqrt{\frac{\bar{p}(1 - \bar{p})}{n_i}} U C L p , i = p ˉ + 3 n i p ˉ ( 1 − p ˉ ) L C L p , i = max ( 0 , p ˉ − 3 p ˉ ( 1 − p ˉ ) n i ) LCL_{p,i} = \max\left(0, \bar{p} - 3 \sqrt{\frac{\bar{p}(1 - \bar{p})}{n_i}}\right) L C L p , i = max ( 0 , p ˉ − 3 n i p ˉ ( 1 − p ˉ ) )
Notice that as n i n_i n i increases, the standard error decreases, causing the control limits to narrow. Conversely, smaller sample sizes cause the limits to expand. Alternatively, engineers construct a Standardized p p p -Chart where the plotted statistic is the standard normal variate Z i Z_i Z i :
Z i = p i − p ˉ p ˉ ( 1 − p ˉ ) n i Z_i = \frac{p_i - \bar{p}}{\sqrt{\frac{\bar{p}(1 - \bar{p})}{n_i}}} Z i = n i p ˉ ( 1 − p ˉ ) p i − p ˉ C L Z = 0 U C L Z = + 3 L C L Z = − 3 CL_Z = 0 \qquad UCL_Z = +3 \qquad LCL_Z = -3 C L Z = 0 U C L Z = + 3 L C L Z = − 3
The n p np n p -Chart (Number Nonconforming)
When operators on a shop floor inspect a fixed number of components each hour (e.g., n = 50 n = 50 n = 50 ), plotting fractional percentages (such as 0.04 0.04 0.04 or 0.06 0.06 0.06 ) can introduce calculation errors. The n p np n p -chart plots the integer count of nonconforming items ( d i d_i d i ) directly.
Important
Strict Operational Requirement: An n p np n p -chart strictly requires a constant subgroup size n n n across all sampling periods. If sample size varies between lots, an n p np n p -chart cannot be used; the engineer must use a p p p -chart.
Formulation for the n p np n p -chart:
C L n p = n p ˉ CL_{np} = n\bar{p} C L n p = n p ˉ U C L n p = n p ˉ + 3 n p ˉ ( 1 − p ˉ ) UCL_{np} = n\bar{p} + 3 \sqrt{n\bar{p}(1 - \bar{p})} U C L n p = n p ˉ + 3 n p ˉ ( 1 − p ˉ ) L C L n p = max ( 0 , n p ˉ − 3 n p ˉ ( 1 − p ˉ ) ) LCL_{np} = \max\left(0, n\bar{p} - 3 \sqrt{n\bar{p}(1 - \bar{p})}\right) L C L n p = max ( 0 , n p ˉ − 3 n p ˉ ( 1 − p ˉ ) )
Where p ˉ = ∑ d i m ⋅ n \bar{p} = \frac{\sum d_i}{m \cdot n} p ˉ = m ⋅ n ∑ d i . Note the mathematical relationship: the n p np n p limits are simply the p p p -chart limits multiplied through by n n n .
- 3. Count of Nonconformities ( c c c ) and Nonconformities per Unit ( u u u ) Charts
- The c c c -Chart (Total Defects per Constant Unit)
The c c c -chart monitors the total count of nonconformities ( c c c ) observed in a single, strictly constant area of opportunity ( n = 1 n = 1 n = 1 ). Typical industrial units include:
Number of weave defects in exactly 100 m 2 100\text{ m}^2 100 m 2 of composite cloth.
Number of solder bridge defects on a single complex avionics circuit board.
Number of surface paint imperfections on an automobile hood.
Given m m m inspection intervals with defect counts c 1 , c 2 , … , c m c_1, c_2, \dots, c_m c 1 , c 2 , … , c m :
c ˉ = 1 m ∑ i = 1 m c i \bar{c} = \frac{1}{m} \sum_{i=1}^m c_i c ˉ = m 1 ∑ i = 1 m c i
Because the variance of a Poisson random variable equals its mean ( σ c 2 = c ˉ ⟹ σ c = c ˉ \sigma_c^2 = \bar{c} \implies \sigma_c = \sqrt{\bar{c}} σ c 2 = c ˉ ⟹ σ c = c ˉ ), the Shewhart 3 σ 3\sigma 3 σ control limits are:
C L c = c ˉ CL_c = \bar{c} C L c = c ˉ U C L c = c ˉ + 3 c ˉ UCL_c = \bar{c} + 3 \sqrt{\bar{c}} U C L c = c ˉ + 3 c ˉ L C L c = max ( 0 , c ˉ − 3 c ˉ ) LCL_c = \max\left(0, \bar{c} - 3 \sqrt{\bar{c}}\right) L C L c = max ( 0 , c ˉ − 3 c ˉ )
The u u u -Chart (Nonconformities per Unit)
When the size of the inspection unit varies between samples (e.g., different rolls of fabric with varying yardage, or inspecting variable batch sizes of electronic subassemblies), the raw defect count c c c cannot be compared directly. The u u u -chart tracks the average defect density per inspection unit ( u = c / n u = c / n u = c / n ):
u i = c i n i u_i = \frac{c_i}{n_i} u i = n i c i
Where n i n_i n i represents the number of inspection units in subgroup i i i , and c i c_i c i is the total count of defects observed across those n i n_i n i units. The grand average defect density across m m m subgroups is:
u ˉ = ∑ i = 1 m c i ∑ i = 1 m n i \bar{u} = \frac{\sum_{i=1}^m c_i}{\sum_{i=1}^m n_i} u ˉ = ∑ i = 1 m n i ∑ i = 1 m c i
Because u i = c i / n i u_i = c_i / n_i u i = c i / n i and Var ( c i ) = n i u ˉ \operatorname{Var}(c_i) = n_i \bar{u} Var ( c i ) = n i u ˉ , the variance of u i u_i u i is Var ( u i ) = n i u ˉ n i 2 = u ˉ n i \operatorname{Var}(u_i) = \frac{n_i \bar{u}}{n_i^2} = \frac{\bar{u}}{n_i} Var ( u i ) = n i 2 n i u ˉ = n i u ˉ . The standard error is σ u , i = u ˉ n i \sigma_{u,i} = \sqrt{\frac{\bar{u}}{n_i}} σ u , i = n i u ˉ .
The control limits for the u u u -chart are:
C L u = u ˉ CL_u = \bar{u} C L u = u ˉ U C L u , i = u ˉ + 3 u ˉ n i UCL_{u,i} = \bar{u} + 3 \sqrt{\frac{\bar{u}}{n_i}} U C L u , i = u ˉ + 3 n i u ˉ L C L u , i = max ( 0 , u ˉ − 3 u ˉ n i ) LCL_{u,i} = \max\left(0, \bar{u} - 3 \sqrt{\frac{\bar{u}}{n_i}}\right) L C L u , i = max ( 0 , u ˉ − 3 n i u ˉ )
If the inspection unit size is constant across all subgroups ( n i = n n_i = n n i = n ), the limits simplify to fixed horizontal lines with denominator u ˉ / n \sqrt{\bar{u}/n} u ˉ / n .
4. Attribute Chart Selection Decision Framework
Selecting the correct control chart is a frequent focus on the PE exam. The decision tree below structures the selection process: Attribute Chart Selection Flowchart
What is being inspected?
- |
- +-----------------------+-----------------------+
- | |
- Nonconforming Units Nonconformities
- (Defective Items) (Defect Counts)
- | |
- Is sample size n Is the area of
strictly constant? opportunity constant?
- | |
- +---+---+ +---+---+
- | | | |
- YES NO YES NO
- | | | |
- np-chart p-chart c-chart u-chart
- (or p) (variable limits) (n = 1) (variable n_i)
- Comprehensive Attribute Comparison Matrix
Attribute Chart Quality Metric Monitored Sample Size ( n n n ) Governing Distribution Center Line ( C L CL C L ) Control Limit Formula ( U C L / L C L UCL / LCL U C L / L C L )
p p p -Chart Fraction Nonconforming Constant or Variable ( n i n_i n i ) Binomial p ˉ \bar{p} p ˉ p ˉ ± 3 p ˉ ( 1 − p ˉ ) n i \bar{p} \pm 3\sqrt{\frac{\bar{p}(1-\bar{p})}{n_i}} p ˉ ± 3 n i p ˉ ( 1 − p ˉ )
n p np n p -Chart Number Nonconforming Strictly Constant ( n n n ) Binomial n p ˉ n\bar{p} n p ˉ n p ˉ ± 3 n p ˉ ( 1 − p ˉ ) n\bar{p} \pm 3\sqrt{n\bar{p}(1-\bar{p})} n p ˉ ± 3 n p ˉ ( 1 − p ˉ )
c c c -Chart Total Defect Count Strictly Constant Area ( n = 1 n=1 n = 1 ) Poisson c ˉ \bar{c} c ˉ c ˉ ± 3 c ˉ \bar{c} \pm 3\sqrt{\bar{c}} c ˉ ± 3 c ˉ
u u u -Chart Defects per Inspection Unit Constant or Variable ( n i n_i n i ) Poisson u ˉ \bar{u} u ˉ u ˉ ± 3 u ˉ n i \bar{u} \pm 3\sqrt{\frac{\bar{u}}{n_i}} u ˉ ± 3 n i u ˉ
5. Lower Control Limit Truncation and Zero-Bound Mechanics
In physical manufacturing systems, the count of defects, count of defective items, and fraction nonconforming have a natural physical boundary at zero ( p ≥ 0 , d ≥ 0 , c ≥ 0 p \ge 0, d \ge 0, c \ge 0 p ≥ 0 , d ≥ 0 , c ≥ 0 ). However, Shewhart formulas calculate symmetrical limits using 3 σ 3\sigma 3 σ margins:
L C L calculated = C L − 3 σ LCL_{\text{calculated}} = CL - 3\sigma L C L calculated = C L − 3 σ
When process quality is high (low p ˉ \bar{p} p ˉ or low c ˉ \bar{c} c ˉ ) or sample size n n n is small, 3 σ 3\sigma 3 σ frequently exceeds the center line value, yielding a negative mathematical result ( L C L < 0 LCL < 0 L C L < 0 ).
Note
The Zero-Truncation Protocol: Whenever the calculated Lower Control Limit is negative, it must be truncated to zero: L C L = 0 LCL = 0 L C L = 0 . Because observed sample values can never fall below zero ( p i ≥ 0 p_i \ge 0 p i ≥ 0 ), a single point can never trigger an out-of-control alarm below the lower control limit on a zero-truncated chart.
Engineering Implications of L C L = 0 LCL = 0 L C L = 0
When L C L = 0 LCL = 0 L C L = 0 , the control chart becomes one-sided for single-point violations. An operational breakthrough (e.g., an improved solder flux yielding zero defects) cannot register as a lower limit breach. To detect significant quality improvements on low-defect lines:
Increase the subgroup size n n n until the calculated L C L LCL L C L is strictly positive ( C L − 3 σ > 0 CL - 3\sigma > 0 C L − 3 σ > 0 ):
- For a p p p -chart: n > 9 ( 1 − p ˉ ) p ˉ n > \frac{9(1 - \bar{p})}{\bar{p}} n > p ˉ 9 ( 1 − p ˉ )
- For a c c c -chart: c ˉ > 9 \bar{c} > 9 c ˉ > 9
Apply run rules (e.g., 8 consecutive points below the center line) to confirm statistically significant quality improvements.
6. Comprehensive Worked Numerical Problem: Multi-Chart Comparison
A quality engineer at an automated electronics manufacturing facility monitors surface-mount technology (SMT) assembly on a high-density telecommunications PCB. The engineer evaluates 25 historical production shifts during stable operations. Across these 25 shifts, exactly n = 100 n = 100 n = 100 PCBs were inspected per shift ( N total = 25 × 100 = 2 , 500 N_{\text{total}} = 25 \times 100 = 2,500 N total = 25 × 100 = 2 , 500 boards).
Inspection records reveal:
A cumulative total of 175 175 175 boards were classified as nonconforming (defective) due to failure on automated optical inspection.
Across all inspected boards, a cumulative total of 625 625 625 solder defects (bridging, insufficient wetting, tombstoning) were recorded. Note that some defective boards contained multiple defects.
Task A: Construct the p p p -Chart and n p np n p -Chart Control Limits
1. Calculate baseline parameters: p ˉ = ∑ d i ∑ n i = 175 2500 = 0.070 \bar{p} = \frac{\sum d_i}{\sum n_i} = \frac{175}{2500} = 0.070 p ˉ = ∑ n i ∑ d i = 2500 175 = 0.070 Subgroup Size: n = 100 \text{Subgroup Size: } n = 100 Subgroup Size: n = 100
2. p p p -Chart Limits: C L p = p ˉ = 0.070 CL_p = \bar{p} = 0.070 C L p = p ˉ = 0.070 σ p = p ˉ ( 1 − p ˉ ) n = 0.070 × ( 1 − 0.070 ) 100 = 0.070 × 0.930 100 = 0.000651 = 0.025515 \sigma_p = \sqrt{\frac{\bar{p}(1 - \bar{p})}{n}} = \sqrt{\frac{0.070 \times (1 - 0.070)}{100}} = \sqrt{\frac{0.070 \times 0.930}{100}} = \sqrt{0.000651} = 0.025515 σ p = n p ˉ ( 1 − p ˉ ) = 100 0.070 × ( 1 − 0.070 ) = 100 0.070 × 0.930 = 0.000651 = 0.025515 3 σ p = 3 × 0.025515 = 0.07654 3\sigma_p = 3 \times 0.025515 = 0.07654 3 σ p = 3 × 0.025515 = 0.07654 U C L p = p ˉ + 3 σ p = 0.070 + 0.07654 = 0.14654 ≈ 0.1465 UCL_p = \bar{p} + 3\sigma_p = 0.070 + 0.07654 = 0.14654 \approx 0.1465 U C L p = p ˉ + 3 σ p = 0.070 + 0.07654 = 0.14654 ≈ 0.1465 L C L p = max ( 0 , 0.070 − 0.07654 ) = max ( 0 , − 0.00654 ) = 0 LCL_p = \max(0, 0.070 - 0.07654) = \max(0, -0.00654) = 0 L C L p = max ( 0 , 0.070 − 0.07654 ) = max ( 0 , − 0.00654 ) = 0
3. n p np n p -Chart Limits (since n = 100 n = 100 n = 100 is constant): C L n p = n p ˉ = 100 × 0.070 = 7.00 defective boards CL_{np} = n\bar{p} = 100 \times 0.070 = 7.00\text{ defective boards} C L n p = n p ˉ = 100 × 0.070 = 7.00 defective boards σ n p = n p ˉ ( 1 − p ˉ ) = 100 × 0.070 × 0.930 = 6.51 = 2.5515 \sigma_{np} = \sqrt{n\bar{p}(1 - \bar{p})} = \sqrt{100 \times 0.070 \times 0.930} = \sqrt{6.51} = 2.5515 σ n p = n p ˉ ( 1 − p ˉ ) = 100 × 0.070 × 0.930 = 6.51 = 2.5515 3 σ n p = 3 × 2.5515 = 7.6545 3\sigma_{np} = 3 \times 2.5515 = 7.6545 3 σ n p = 3 × 2.5515 = 7.6545 U C L n p = 7.00 + 7.6545 = 14.6545 ≈ 14.65 boards UCL_{np} = 7.00 + 7.6545 = 14.6545 \approx 14.65\text{ boards} U C L n p = 7.00 + 7.6545 = 14.6545 ≈ 14.65 boards L C L n p = max ( 0 , 7.00 − 7.6545 ) = 0 boards LCL_{np} = \max(0, 7.00 - 7.6545) = 0\text{ boards} L C L n p = max ( 0 , 7.00 − 7.6545 ) = 0 boards
Notice that U C L n p = n × U C L p = 100 × 0.14654 = 14.654 UCL_{np} = n \times UCL_p = 100 \times 0.14654 = 14.654 U C L n p = n × U C L p = 100 × 0.14654 = 14.654 boards.
Task B: Construct the u u u -Chart Limits
Because the engineer is tracking the count of nonconformities (defects) across inspection units of n = 100 n = 100 n = 100 boards per shift, a u u u -chart monitors the average defects per board:
1. Calculate baseline defect rate per board ( u ˉ \bar{u} u ˉ ): u ˉ = ∑ c i ∑ n i = 625 2500 = 0.250 defects per board \bar{u} = \frac{\sum c_i}{\sum n_i} = \frac{625}{2500} = 0.250\text{ defects per board} u ˉ = ∑ n i ∑ c i = 2500 625 = 0.250 defects per board
2. u u u -Chart Limits for standard shift ( n = 100 n = 100 n = 100 boards): C L u = u ˉ = 0.250 defects/board CL_u = \bar{u} = 0.250\text{ defects/board} C L u = u ˉ = 0.250 defects/board σ u = u ˉ n = 0.250 100 = 0.0025 = 0.050 defects/board \sigma_u = \sqrt{\frac{\bar{u}}{n}} = \sqrt{\frac{0.250}{100}} = \sqrt{0.0025} = 0.050\text{ defects/board} σ u = n u ˉ = 100 0.250 = 0.0025 = 0.050 defects/board 3 σ u = 3 × 0.050 = 0.150 defects/board 3\sigma_u = 3 \times 0.050 = 0.150\text{ defects/board} 3 σ u = 3 × 0.050 = 0.150 defects/board U C L u = u ˉ + 3 σ u = 0.250 + 0.150 = 0.400 defects/board UCL_u = \bar{u} + 3\sigma_u = 0.250 + 0.150 = 0.400\text{ defects/board} U C L u = u ˉ + 3 σ u = 0.250 + 0.150 = 0.400 defects/board L C L u = u ˉ − 3 σ u = 0.250 − 0.150 = 0.100 defects/board LCL_u = \bar{u} - 3\sigma_u = 0.250 - 0.150 = 0.100\text{ defects/board} L C L u = u ˉ − 3 σ u = 0.250 − 0.150 = 0.100 defects/board
Notice that because σ u \sigma_u σ u is small, the lower control limit is strictly positive ( L C L u = 0.100 > 0 LCL_u = 0.100 > 0 L C L u = 0.100 > 0 ). If a shift exhibits an average defect rate below 0.100 0.100 0.100 defects/board, it triggers an assignable cause alarm indicating a statistically significant quality improvement!
3. Evaluation of an Unusual Batch: On shift 26, line maintenance reduces production such that only n 26 = 36 n_{26} = 36 n 26 = 36 boards are assembled. Inspectors record 12 12 12 total defects on these 36 boards. Is Shift 26 in statistical control?
Observed rate: u 26 = 12 / 36 = 0.3333 defects/board u_{26} = 12 / 36 = 0.3333\text{ defects/board} u 26 = 12/36 = 0.3333 defects/board .
Revised standard error for n 26 = 36 n_{26} = 36 n 26 = 36 : σ u , 26 = u ˉ n 26 = 0.250 36 = 0.500 6 = 0.08333 defects/board \sigma_{u,26} = \sqrt{\frac{\bar{u}}{n_{26}}} = \sqrt{\frac{0.250}{36}} = \frac{0.500}{6} = 0.08333\text{ defects/board} σ u , 26 = n 26 u ˉ = 36 0.250 = 6 0.500 = 0.08333 defects/board U C L u , 26 = 0.250 + 3 ( 0.08333 ) = 0.250 + 0.250 = 0.500 defects/board UCL_{u,26} = 0.250 + 3(0.08333) = 0.250 + 0.250 = 0.500\text{ defects/board} U C L u , 26 = 0.250 + 3 ( 0.08333 ) = 0.250 + 0.250 = 0.500 defects/board L C L u , 26 = max ( 0 , 0.250 − 0.250 ) = 0 defects/board LCL_{u,26} = \max(0, 0.250 - 0.250) = 0\text{ defects/board} L C L u , 26 = max ( 0 , 0.250 − 0.250 ) = 0 defects/board
Since u 26 = 0.3333 u_{26} = 0.3333 u 26 = 0.3333 falls well within [ 0 , 0.500 ] [0, 0.500] [ 0 , 0.500 ] , Shift 26 remains in statistical control despite the smaller sample size.
7. Common PE Exam Pitfalls for Attribute Charts
Using an n p np n p -Chart with Variable Subgroup Sizes: An n p np n p -chart tracks the count of defectives, which directly scales with sample size. If an exam problem states that lot sizes fluctuate from 80 to 120 units, selecting an n p np n p -chart is an immediate disqualifier; you must use a p p p -chart.
Confusing Defect Counts with Defective Unit Counts: A part with 4 solder bridges is 1 defective unit, but contributes 4 defects. Watch the problem statement terminology carefully: "fraction nonconforming" or "defective items" dictates p p p or n p np n p ; "defects per unit" or "imperfections per roll" dictates c c c or u u u .
Failing to Truncate Negative Lower Control Limits: When calculations yield L C L = − 0.024 LCL = -0.024 L C L = − 0.024 , never select an answer option displaying a negative limit. The lower limit for attribute charts is bounded by zero ( L C L = 0 LCL = 0 L C L = 0 ).
Forgetting n \sqrt{n} n in p p p -Chart Standard Error: The standard deviation of the proportion is p ˉ ( 1 − p ˉ ) n \sqrt{\frac{\bar{p}(1-\bar{p})}{n}} n p ˉ ( 1 − p ˉ ) . Examinees occasionally forget to divide by n n n inside the radical or erroneously multiply by n n n , confusing σ p \sigma_p σ p with σ n p \sigma_{np} σ n p .