https://widener.instructure.com/courses/28935/pages/week-5-learning-resourcesWatch videos on link attached above and attachments to complete a control chart using Excel.
Control Chart Assignment
1. Data for cycle time in minutes is given in “Control Chart Assignment Data.xlsx”. Use
Data Set A. The data contains: 36 daily sample means (X-bar) of cycle time. With this data construct a control chart (note: there are a variety of possible control charts; for this assignment use an X-bar Chart)
using 2 times the standard deviation
as the control limits
(if you Google Control Charts, you will find references to 3 times standard deviations – in this example, 2 works much better and is why we are directing you to use 2). Analyze the control chart and discuss whether or not the process is “under control”. Observe the data and discuss if it suggests the existence of any other problems.
2.
Data Set B
pertains to a period after a process improvement project was completed. Develop a control chart for this data similar to what you developed for Data Set A. Considering this data in comparison to Data Set A, was the project successful? Justify your answer as thoroughly as possible.
Note: The control charts should be constructed in Excel and the discussion should be submitted in a Word file. Please submit both files.
Control Chart Assignment 1. Data for cycle time in minutes is given in “Control Chart Assignment Data.xlsx”. Use Data Set A. The data contains: 36 daily sample means (X-bar) of cycle time. With this data construct a control chart (note: there are a variety of possible control charts; for this assignment use an X-bar Chart) using 2 times the standard deviation as the control limits (if you Google Control Charts, you will find references to 3 times standard deviations – in this example, 2 works much better and is why we are directing you to use 2). Analyze the control chart and discuss whether or not the process is “under control”. Observe the data and discuss if it suggests the existence of any other problems. 2. Data Set B pertains to a period after a process improvement project was completed. Develop a control chart for this data similar to what you developed for Data Set A. Considering this data in comparison to Data Set A, was the project successful? Justify your answer as thoroughly as possible. Note: The control charts should be constructed in Excel and the discussion should be submitted in a Word file. Please submit both files. How to create a Control Chart in Excel 1. Open “Control Chart Assignment Data.xlxs”. There are two tabs labeled Data Set A and Data Set B respectively 2. Label cells N1 N2,N3 & N4 “Mean”,”UCL” & “LCL” , respectively. 3. In cell O1, type =AVERAGE(B2:B37) to calculate the mean 4. In cell O2, type =STDEV.S(B2:B37) to calculate the standard deviation(note STDEV.S is used because data comes from samples) 5. In cell O3, type =(O1+(2*O$2)) to calculate the UCL 6. In Cell O4, type =(O1-(2*O$2)) to calculate the UCL 7. Enter the values you calculated for MEAN, UCL and LCL in columns C D and E. Your spreadsheet should look like Figure 1 below Figure 1 Spreadsheet for Control Chart Data Now you are ready to construct the control chart in Excel (this requires the creation of a line chart). 1. In your data area, highlight Daily X-bar, Mean, UCL & LCL and all the data under these column labels(this will create a line for each of these) 2. Select insert 3. Select line 4. Select 2D 5. Highlight a data point and right click it 6. Select format data series 7. Select Marker options 8. Select built in (this inserts a square for each data point). You now have a control chart that looks like Figure 2. You can also see a completed control chart in your Excel file. Figure 2 Control Chart for Cycle Time 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Daily X-Bar Daily X-bar Mean UCL LCL 0 10 20 30 40 50 60 70 80 90 100 123456789101112131415161718192021222324252627282930313233343536 Daily X-Bar Daily X-barMeanUCLLCL Data Set A Sample #Daily X-bar Vertex42: X-bar is the sample mean calculated as the sum of the observations divided by the number of observations in the sample (n).MeanUCLLCL# Special Orders 1230 2432 3230 4351 5281 6392 7302 8895 9251 10271 11292 12282 13330 14353 15362 16381 17420 18340 19320 20300 21280 22453 23322 24280 25290 26392 27886 28341 29280 30392 31371 32361 33422 34402 35371 36895 Data Set B Sample #Daily X-bar Vertex42: X-bar is the sample mean calculated as the sum of the observations divided by the number of observations in the sample (n).MeanUCLLCL# Special Orders 1250 2452 3270 4321 5271 6352 7282 8475 9221 10211 11302 12312 13260 14413 15372 16351 17340 18370 19390 20320 21330 22443 23322 24260 25220 26352 27526 28321 29290 30322 31351 32371 33412 34392 35361 36485 Introduction to Process Improvement & Managing Process Improvements Study, Test, Monitor: Statistical Process Control BUS665 – Managing Business Processes 1 Agenda Part I Understanding variation What should we measure? Tools – run & control charts How to interpret Part II Types of control charts Examples 2 Part I What is Variation? Measures Interpreting Run & Control Charts 3 Let’s recall… A well-defined process contains five core components… 4 Resources: the things a process requires to routinely convert inputs to outputs Tangible: people, computer, software Inputs: the things that are transformed by the process into an end product or service Tangible: written data, parts, and forms Intangible: verbal requests Activities: actions that move the inputs through the process to become outputs Tangible: measuring, sawing, nailing, painting, and writing Intangible: reading, approving, or submitting Outputs: products, services, or information Tangible: products Intangible: advice Controls: activities involved in ensuring a process is predictable and stable Internal: organization standards External: customer specifications, legislative requirements, copyright laws What is Variation? All processes have outputs… We can (and should) measure these outputs to ensure they meet our standards… The distribution of these measures will vary…these differences = process variation Shewhart (in 1924, the father of SPC) discovered some variation is inherent in all systems Some is going to be acceptable (within our predetermined allowable limits)—COMMON CAUSE Results from the system – variation caused by hiring, training, and supervisory policies and practices—RESPONSIBILITY OF MANAGEMENT Some is going to be unacceptable—SPECIAL CAUSE Fluctuations or patterns not inherent to a process—RESPONSIBILITY OF FRONTLINE WORKERS Difficult to design a perfect system (that creates absolutely perfect outputs 100% of the time) so we aim for one that is in statistical control – no special cause variation UNLESS – we’re improving a process and hope to see our measures arrive at a new normal Think about healthcare - the goal is 0 harm! Special cause variation in this case might tell us our intervention is working! 5 What should we measure? 6 DEPENDS! Going to vary by company; each needs to determine what should be measured – what would provide valuable information on how we’re achieving our mission? Can use a variety of perspectives – operational, financial, customer, quality, etc.… All of these could be graphed on run charts or control charts to see if they are stable, improved, or worsened Operational Transaction volume Cycle time Financial Cost of each subprocess – staff, material Cost of goods sold – raw materials/labor Customer Experience Quality Backlogs of subprocesses Number of errors in processing; defects Waste – time, material Frequency and number of delays Scrap Customer complaints Customer interaction – w/CSRs, web portal, phone Problem resolution Order processing errors Six Sigma, TQM measures Audit/inspection of final product –error and rejection What should we measure? In healthcare we specifically look at the following classifications of measures… 7 Provide a sense of an organization’s capacity, systems, and processes to provide high-quality care Indicate what a provider does to maintain or improve health – typically reflect standards of care Reflect the impact of the healthcare service or intervention on the health status of patients Structural Does the organization use EHR Number or proportion of board-certified physicians Ratio of providers to patients Process Percentage of people receiving preventative services ( such as mammograms or immunizations) Percentage of people with diabetes who had their blood glucose tested and controlled Outcome Percentage of patients who died as a result of surgery (surgical mortality rates) The rate of surgical complications or HAIs Acute myocardial infarction (AMI) patients without aspirin contraindications who received aspirin within 24 hours before or after hospital arrival Measuring Variation How do we find out of if we have special or common cause variation in our system? Basic run charts A series of rules to determine meaningful patterns Control charts Attribute P charts C charts U charts Variables Individuals and moving range (I-MR) charts X Bar and R charts X Bar and S charts 8 Interpreting Run Charts FOUR RULES The presence of any single rule is evidence of a non-random signal of change (there is less than 5% likelihood that the conditions of the rule will be met simply by chance) 9 Interpreting Run Charts 10 Guide available here Creating a Run Chart in Minitab Let’s do a quick example… [TRANSMIT] dataset 11 Creating a Run Chart in Minitab 12 Control Charts Attempt to separate special or assignable causes of variation from random noise or common cause variation Special cause = a change NOT attributable to the system BAD if we haven’t intentionally introduced a change GOOD if we have just engaged in a process improvement project All control charts have a common structure… Measurement of a process characteristic, plotted against time (or other variable – employee or location) A central line that represents the process average (mean) Upper (UCL) and lower (LCL) control limits – adding and subtracting three standard deviations from the process mean 13 An Example Control Chart… 14 Interpreting a Control Chart Once control limits are calculated, the chart is evaluated for any nonrandom patterns that might exist… 15 Panel A: No discernable pattern in the ordering of values over time, and no points fall outside the control limits STABLE, only common cause variation Panel B: Two points fall outside the control limits – each should be investigated to determine the causes that led to their occurrence UNSTABLE, special cause variation is present Panel C: No points fall outside the control limits but it has a series of consecutive points above and below the average value (center line), and a long-term downward trend in the value of the variable UNSTABLE, special cause variation is present Control limits often called three-sigma limits… A six sigma process is one in which 99.99966% of all opportunities to produce some an output are statistically expected to be free of defects Interpreting Control Charts Other zone boundaries can be helpful in detecting other unlikely patterns… 16 68.26% within this range 95.44% within this range 99.73% within this range RULES Rule 1—A point falls above the UCL or below the LCL Due to 1 of 2 possibilities… A special cause occurred A common cause occurred with the likelihood of 0.135% or 1,350 times per million opportunities 17 Interpreting Control Charts RULES Rule 2—Eight or more consecutive points lie above or below the center line 18 Interpreting Control Charts RULES Rule 3—Six or more consecutive points move upward or downward in value 19 Interpreting Control Charts RULES Rule 4—An unusually small number of consecutive points above and below the center line (saw tooth pattern) Is a process unusually noisy (high variability) or unusually quiet (low variability)? 20 Interpreting Control Charts RULES Rule 5—Two out of three consecutive points fall in the high Zone A or above, or in the low Zone A or below 21 Interpreting Control Charts RULES Rule 6—Four out of five consecutive points fall in the high Zone B or above, or in the low Zone B or below 22 Interpreting Control Charts RULES Rule 7—Fifteen consecutive points fall within Zone C on either side of the center line Is a process unusually noisy (high variability) or unusually quiet (low variability)? 23 Interpreting Control Charts Interpreting Control Charts If only common causes are operating, each of the preceding 7 patterns are statistically UNLIKELY to occur special cause variation is present 24 Interpreting a Control Chart A quick