UPSC MainsMANAGEMENT-PAPER-II20215 Marks
Q20.

Statistical Process Control: Control Charts

A manufacturing company always uses a sample size of 10. Data for the past 10 samples are as follows : Number of Defective Units Total Units Sampled 150 1,000 The system is believed to be operating under normal conditions. (A) Construct a control chart with control limits such that measurements for 95% of the units under normal conditions would fall within the control limits. (B) Suppose the number of units defective in the next 5 samples are: 3, 4, 2, 0 and 7. Demonstrate graphically whether the process is under control or not.

How to Approach

This question tests the understanding of Statistical Process Control (SPC) and the application of control charts. The approach should involve calculating the control limits (UCL and LCL) for a p-chart (proportion defective) based on the given sample data. Then, plotting the subsequent sample data on the chart and determining if the process is in control by observing if any points fall outside the control limits. The answer should demonstrate a clear understanding of the underlying principles and calculations.

Model Answer

0 min read

Introduction

Statistical Process Control (SPC) is a method of quality control which employs statistical methods to monitor and control a process. Control charts are a key tool in SPC, visually displaying process data over time and helping to identify variations that may indicate a process is out of control. A p-chart, specifically, is used to monitor the proportion of defective items in a sample. Establishing control limits, typically at ±3 standard deviations from the mean, allows for the identification of assignable causes of variation and ensures process stability. This question requires the construction and interpretation of a p-chart to assess the stability of a manufacturing process.

(A) Constructing the Control Chart

The given data represents the number of defective units and the total units sampled for 10 samples. To construct a p-chart, we need to calculate the sample proportion defective (p̄) and the control limits (UCL and LCL).

1. Calculate the Sample Proportion Defective (pi) for each sample:

pi = (Number of Defective Units) / (Total Units Sampled)

In this case, pi = 150/1000 = 0.15 for all 10 samples as the data is the same for all samples.

2. Calculate the Average Proportion Defective (p̄):

p̄ = (Σpi) / n, where n is the number of samples.

p̄ = (10 * 0.15) / 10 = 0.15

3. Calculate the Standard Deviation of the Sample Proportions (σp):

σp = √[p̄(1-p̄)/n]

σp = √[0.15(1-0.15)/10] = √[0.15 * 0.85 / 10] = √0.01275 ≈ 0.1129

4. Calculate the Control Limits (UCL and LCL):

UCL = p̄ + 3σp = 0.15 + 3 * 0.1129 = 0.15 + 0.3387 = 0.4887

LCL = p̄ - 3σp = 0.15 - 3 * 0.1129 = 0.15 - 0.3387 = -0.1887

Since the LCL cannot be negative, it is set to 0.

Therefore, the control chart has a UCL of 0.4887 and an LCL of 0.

(B) Demonstrating Process Control

Now, let's plot the proportion defective for the next 5 samples (3, 4, 2, 0, and 7 defective units, assuming a constant sample size of 1000) and see if the process is under control.

1. Calculate the Proportion Defective for the next 5 samples:

  • Sample 11: p11 = 3/1000 = 0.003
  • Sample 12: p12 = 4/1000 = 0.004
  • Sample 13: p13 = 2/1000 = 0.002
  • Sample 14: p14 = 0/1000 = 0.000
  • Sample 15: p15 = 7/1000 = 0.007

2. Graphical Representation:

Imagine a graph with sample number on the x-axis and proportion defective on the y-axis. The UCL is at 0.4887, the center line (p̄) is at 0.15, and the LCL is at 0. All five points (0.003, 0.004, 0.002, 0.000, 0.007) fall well within the control limits (0 to 0.4887).

Conclusion: Since none of the points fall outside the control limits, the process is considered to be under control. The observed variations are likely due to random chance and are within the expected range for a stable process.

Conclusion

In conclusion, we successfully constructed a p-chart with UCL at 0.4887 and LCL at 0, based on the initial data. The subsequent analysis of the next five samples demonstrated that all data points fell within the established control limits, indicating that the manufacturing process remains stable and under statistical control. Continuous monitoring using control charts is crucial for maintaining quality and identifying potential issues before they escalate.

Answer Length

This is a comprehensive model answer for learning purposes and may exceed the word limit. In the exam, always adhere to the prescribed word count.

Additional Resources

Key Definitions

Statistical Process Control (SPC)
SPC is the use of statistical methods to monitor and control a process. It helps to ensure that the process operates efficiently, producing more specification-conforming products with less waste.
Assignable Cause
An assignable cause is a specific, identifiable reason for variation in a process. These causes are not inherent to the process and can be eliminated to improve process stability.

Key Statistics

According to a study by the American Society for Quality (ASQ), companies that implement SPC experience an average reduction of 25-50% in defects.

Source: American Society for Quality (ASQ), 2023 (Knowledge Cutoff)

A 2021 report by Statista indicated that the global market for Statistical Process Control software was valued at approximately $250 million.

Source: Statista, 2021 (Knowledge Cutoff)

Examples

Toyota Production System

Toyota's renowned Production System heavily relies on SPC principles, particularly control charts, to monitor and improve manufacturing processes, leading to high quality and efficiency.

Frequently Asked Questions

What is the significance of the 3-sigma control limits?

The 3-sigma control limits are based on the empirical rule (68-95-99.7 rule) in statistics, which states that approximately 99.7% of data points will fall within 3 standard deviations of the mean in a normal distribution. This means that only about 0.3% of points are expected to fall outside the limits due to random variation.

Topics Covered

StatisticsOperations ManagementControl ChartsQuality ControlProcess Analysis