Statistical Process Control (SPC) Bootcamp (Accredited)

Explore basic statistics behind SPC, including data types, central tendency, and dispersion. Learn normal, binomial, and Poisson distributions, plus charts such as x-bar and p charts.

Explore data types by distinguishing discrete and continuous data, classify data as nominal, ordinal, interval, or ratio, and relate them to control charts, histograms, and bar charts.

Learn how to summarize data using central tendency by computing the mean, median, and mode, and understand how these measures relate to dispersion and histogram peaks.

Explore the three measures of dispersion—range, standard deviation, and variance—and how they describe data spread, including population versus sample standard deviation and their role in control charts.

Learn how to compute descriptive statistics in Excel, using the Analysis Toolpak and formulas to obtain mean, median, mode, standard deviation, variance, range, min, max, and count from data sets.

Learn multiple methods to estimate standard deviation for control charts, including range/4 and R-bar over D2, and recognize population vs. sample standard deviation and distribution-based formulas.

Plot discrete data with bar charts and continuous data with histograms or line charts to visualize patterns, and build Excel-based control charts with an average line and control limits.

Explore probability distributions in SPC: use the normal distribution for continuous data and binomial or Poisson for discrete data in control charts.

Develop understanding of the binomial distribution in SPC, covering probability concepts, two-outcome trials, independence, and the binomial probability formula using n and x.

Master binomial distribution with Excel demonstrations using BINOM.DIST to compute exact and cumulative probabilities, and derive mean and standard deviation for p-chart and np-chart applications.

Compare binomial and Poisson distributions for discrete data, noting that binomial limits successes while Poisson handles infinite possibilities. Use Excel to compute probabilities of defects, road accidents, and queue lengths.

Explore Poisson distribution calculations with a mean of 3.6 per 10 minutes to find the probability of 7 events and learn how mu equals the variance for control charts.

Understand the normal distribution for continuous data, its symmetry and mean–mode–median equality, and how mean and standard deviation define probabilities within 1–3 standard deviations for SPC control limits.

Examine the normal distribution and standard normal form, convert values to z-scores, and find probabilities using standard normal tables or Excel, focusing on area under the curve.

Learn to calculate areas under the standard normal distribution with Excel using norm.s.dist, and apply standard deviation formulas for x-bar s charts, imr and x-bar r charts, and cp/cpk.

Understand how the central limit theorem makes sample means form a normal distribution with larger n, enabling normal-based control charts.

Define subgroup concepts in X bar and R charts, using five items. Differentiate within-subgroup from between-subgroup variation and relate to how control limits form.

Explore type I and type II errors in control charts. Distinguish false alarms from missed process changes and learn how subgroup size influences these errors.

Explore control charts basics, including seven common charts for attribute and variable data, with cusum and ewma, and practice creating control limits in Excel and Minitab.

Distinguish control limits from specification limits using an x-bar chart, showing how control limits reflect process behavior and specification limits express customer requirements.

Learn the two phases of control charts: setting initial upper and lower control limits from samples, then using fixed limits for ongoing process control, with root cause analysis for outliers.

Choose the right control chart by data type: discrete data use Np, P, C, or U charts; variable data use X Bar R, or X Bar S charts.

Build an Np chart in Excel and Minitab by calculating Np bar, P bar, and control limits, then interpret the defectives data on a control chart.

Explore p chart theory for the proportion of defectives with variable subgroup sizes, with varying control limits, p-bar calculation, and practical Excel and Minitab demonstrations.

Learn to construct a P chart for proportion defective using Excel and Minitab, calculating P-bar, upper and lower control limits, and applying it to varying subgroup sizes.

Study c chart and u chart for defect counts using Poisson distribution, with constant or variable subgroup sizes, and compute control limits with C bar and standard deviation sqrt(C bar).

Learn to build a C chart in Excel and Minitab, compute the average defects (C bar), and set upper and lower control limits to assess defect counts.

Explore the U chart theory for attributes when subgroup size varies, using proportion defects and U bar, with changing control limits and examples in Excel and Minitab.

Build a u chart in Excel and Minitab from units inspected and defects data, compute u bar, and set dynamic upper and lower control limits for each reading.

Learn to use X bar R, X bar S, and IMR charts for variable data, select by subgroup size, and set control limits with A2, D3, D4.

Learn to build X bar and R charts with Excel and Minitab, compute means and ranges, determine control limits using X bar bar, R bar, A2, D3, and D4, and interpret subgroup variation.

Learn the x bar s chart for variable data when subgroups exceed nine, using standard deviation and control limits with A3, B3, and B4, plus plotting in Excel and Minitab.

Learn to build an x bar s chart using Excel and Minitab, compute x bar and s for 10-item subgroups, and establish upper and lower control limits.

Explore the IMR (XMR) chart, its moving range between consecutive items with subgroup size 2, and compute control limits using D3 and D4 in Excel and Minitab.

Learn to construct an I-MR (XMR) chart from PH measurements using Excel and Minitab, calculating moving ranges, MR bar, and control limits for individuals and ranges.

Explore the Nelson rules for control charts, detailing eight criteria beyond a single out-of-control point, including nine consecutive points on one side, with upcoming coverage of the first four rules.

Learn Nelson rules: rule 1 flags out-of-control when a point exceeds ±3 standard deviations; nine points on one side, six consecutive points forming a trend, and fourteen alternating indicate issues.

Explore Nelson rules 5–8 for control charts. Identify patterns beyond 2 standard deviations or 1 standard deviation, 15 in a row, or 8 in a row outside, signaling process change.

Discover pre-control charts, using specification limits to set upper and lower pre-control limits. Apply a simple two-item green zone rule to decide when a process is ready to start.

Learn to create short run charts with difference and IMR charts, and ZMR charts using Excel and Minitab, including moving range, nominal value difference, and standard deviation calculations.

Discover how EWMA and CUSUM time-weighted control charts detect very small shifts in the process mean using lambda weighting and moving averages.

Demonstrates creating EWMA charts in Excel and Minitab using a moving average of 2 and lambda 0.2 to detect small mean shifts, with upper and lower control limits.

Explore the cusum control chart, its upper and lower plots, and the tabular approach with k and h values for detecting small process mean shifts, using Excel and Minitab.

Demonstrates constructing a CUSUM chart from pigment data in Minitab and Excel, using the tabular approach, with H=4, K=0.5, target 10, to detect shifts.

Explore process capability by learning Cp, Cpk, Pp, and Ppk to determine if a process meets expectations, including normal and non-normal data considerations.

Explore how control charts assess process stability and capability using I and MR charts, UCL and LCL, and specification limits, with Cp, Cpk, Pp, and Ppk.

Explore how to link control charts and histograms using PH value data, demonstrating how process capability is assessed with Cp, Cpk, Pp, and Ppk.

Introduce Cp, Cpk, Pp, and Ppk and their calculation under four conditions: representative samples, normal distribution, statistical control via a control chart, and sufficient sample size.

Demonstrate calculating Cp and related metrics (Cpk, Pp, Ppk) in Minitab from 75 coating-thickness readings, using sigma within and USL/LSL, plus an individual value plot.

Explore the difference between sigma within and sigma overall, and how Cp, Cpk, Pp, and Ppk relate to short-term versus long-term process capability, using Rbar, D2, and Excel/Minitab.

Calculate Cp in Excel for SPC by deriving sigma within from R bar and D2, using USL 53 and LSL 47, with a 3-measurement subgroup example.

Calculate Cp as (upper minus lower specification) divided by six sigma within, using sigma within from Rbar/D2. Both machines yield Cp about 2.42, yet machine 2’s mean shifts, highlighting Cp's limitation.

Compare Cp and Cpk calculations using two machines with the same standard deviation but different means; Cp remains 2.42 while Cpk reveals centering, yielding 2.32 and 0.90.

Compare Cp, Cpk with Pp, Ppk, showing how long-term standard deviation affects Pp and Ppk through Excel and Minitab demonstrations. Emphasize asking suppliers for Ppk to assess overall process capability.

Assess normality before capability analysis for Cp, Cpk, Pp, Ppk using histograms, QQ plots, and Anderson-Darling tests, interpret p-values, and see practical demos in Excel and Minitab.

Learn to assess normality using Excel histograms with adjustable bins and confirm with Minitab's Anderson-Darling test and QQ or probability plots, interpreting p-values.

Explore how to assess non-normal data using Minitab to identify a best-fit distribution (weibull) and compute process capability indices (Pp, Ppk) from tile warping data, revealing a not capable process.

The lecture guides normality checks with a normality test, identifies the Weibull distribution for non-normal data, and performs a process capability analysis with a 0 to 6 limit and Ppk.