Control charts are statistical tools used to monitor and control a process. They graphically display data over time to detect unusual patterns or variations. The document discusses the theory and construction of different types of control charts including R, P, and C charts. Control charts can identify when a process is out of control so that corrective actions can be taken to improve quality.
Control charts are statistical tools used to monitor and control a process. They graphically display data over time to detect unusual patterns or variations. The document discusses the theory and construction of different types of control charts including R, P, and C charts. Control charts can identify when a process is out of control so that corrective actions can be taken to improve quality.
Control charts are statistical tools used to monitor and control a process. They graphically display data over time to detect unusual patterns or variations. The document discusses the theory and construction of different types of control charts including R, P, and C charts. Control charts can identify when a process is out of control so that corrective actions can be taken to improve quality.
Control charts are statistical tools used to monitor and control a process. They graphically display data over time to detect unusual patterns or variations. The document discusses the theory and construction of different types of control charts including R, P, and C charts. Control charts can identify when a process is out of control so that corrective actions can be taken to improve quality.
Shewart, are also known as Shewart charts • It is a statistical technique for controlling the quality of a product being manufactured • It can predict the rejects when they are likely to occur, which enables corrective action to be taken before a defective product is actually produced Theory of control charts • It is well recognized that certain variations in the quality of any product will always occur • Natural variations follow the normal curve • When the variations follow the normal curve, the process is under control, otherwise, out-of-control • If process is out-of-control, it is required to identify the cause for the variation and try to eliminate it » Continued….. Theory of control charts • The maximum variation in the quality characteristic is + 3 σ, i.e., the variation in the quality characteristic of a product should not exceed + 3 σ from the mean value • Control chart is drawn by plotting mean value(μ), the upper control limit(μ+ 3 σ), and the lower control limit(μ- 3 σ) • By plotting the measurements of the quality characteristic of the products over the control chart, if these lies in between the upper control limit(μ+ 3 σ), and the lower control limit(μ- 3 σ) then the process is in-control, otherwise, out-of-control Range control charts (R charts) • These are used for controlling the range or dispersion of the process producing a product • Upper control limit = μ+3 σ, lower control limit = μ-3 σ, where μ is mean of the sample ranges and σ is standard deviation of sample ranges • σ = (d3/d2)* μ, d3 and d2 vary with the number of observations in a sample and their values are given in a table » Continued….. Range control charts (R charts) • Upper control limit = μ+3 σ =(1+3d3/d2)* μ =D4* μ • lower control limit = μ-3 σ =(1-3d3/d2)* μ =D3* μ • The values of D3 and D4 are also given in the table Construction of R charts • The range for a sample is determined as subtracting the lower value from the higher one • Determine the mean of the sample ranges • Draw the mean value over the control chart • Determine the upper control limit and the lower control limit and draw them • Plot the samples ranges over the R chart Analysis of R chart • If all the points are within the UCL and the LCL then the process is in-control , otherwise, out- of-control – If some of the points are lying outside the control limits then it is required to identify the cause for them and try to eliminate – In case if some observation is outside the LCL then it would be good to find the cause for this observation Process capability study • Process capability is the quality performance capability of a process under in control conditions – It is defined as 6 σ, where σ = μ / d2 – Process capability = 6 * μ /d2 • Process capability study is to assess whether the process can meet the job tolerance limit for its long run – job tolerance limit is specified with the problem – process capability should be less than the job tolerance limit Use of control charts • The control chart can be placed near the process so that the operator himself can check the quality of the product – By plotting the observed sample values over the chart , it can be decided whether the process is in-control or out-of-control – For the out-of-control process, cause should be identified and it should be eliminated • When should go for sampling it depends on the speed of the process – In a process with high rate of production and rapid process drift, the sampling should be done at shorter intervals as compared to a stable process Control chart of attributes • Control chart of attributes are used in a situation where the quality characteristics may not be quantified • The term defect refers to the failure of a quality characteristic to meet the specified standard • The term defective designates an item with one or more defects » Continued…… Control chart of attributes
• The quality of a product inspected for
attributes can be expressed as fraction defective or defect per unit – fraction defective= number of units found defective / total number of units inspected – defect per unit= total number of defects observed / number of units inspected Control charts for fraction defective • These are known as P charts, where P stands for fraction defective • Control limits for the P charts: (1 ) UCL 3 3 N CENTRE LINE (1 ) LCL 3 3 ,Where refers to average fraction defective, N N refers to sample size, and refers to s tan dard deviation Construction of P charts • The fraction defective for a sample is determined • Determine the mean of the samples fraction defective • Draw the mean value over the control chart • Determine the upper control limit and the lower control limit and draw them • Plot the samples ranges over the P chart Analysis of P chart • If all the points are within the UCL and the LCL then the process is in-control , otherwise, out- of-control – If some of the points are lying outside the control limits then it is required to identify the causes for them and try to eliminate them by taking some action so that the quality can be improved – In case if some observation is outside the LCL then • either check the effectiveness of the inspection or find the cause for this observation P charts with variable sample size • In many cases, P charts are based on 100% inspection of output of a process which may be varying at the time of inspection • In order to draw the control charts, use the average of the sample sizes as sample size to determining the UCL and the LCL • Plot the samples over the P chart • If for some of the samples, points lie beyond the UCL and the LCL, then use their sample sizes in order to determine the UCL and the LCL to assess whether the process is in control or out of control Number of defects control charts (C charts)
• C charts are used for such types of products
where some defects are always found • Control limits for the C charts: UCL 3 3 CENTRE LINE LCL 3 3 , Where refers to average number of defects per unit , and refers to s tan dard deviation Construction of C charts • The product is inspected to determining the number of defects • Determine the mean of the number of defects per unit • Draw the mean value over the control chart • Determine the upper control limit and the lower control limit and draw them • Plot the number of defects per unit over the P chart Analysis of C charts • If all the points are within the UCL and the LCL then the process is in-control , otherwise, out- of-control – If some of the points are lying outside the control limits then it is required to identify the causes for them and try to eliminate them by taking some action so that the quality can be improved – In case if some observation is outside the LCL then • either check the effectiveness of the inspection or find the cause for this observation