Continuous Improvement - PDCA ( Plan-do-dheck-action)

Views:
 
     
 

Presentation Description

Problem solving tools and self improvement through PDCA cycle

Comments

Presentation Transcript

Continuous Improvement:

1 Continuous Improvement Moutaz Khouja Business Information Systems and Operations Management Department The Belk College of Business Administration The University of North Carolina at Charlotte Charlotte, NC 28223 Ph: (704) 687-3242, Fax: (704) 687-6330 E-mail: mjkhouja@email.uncc.edu

Continuous Improvement:

2 Continuous Improvement Overview of the PDCA Problem Solving Cycle. Foundations of the PDCA Cycle Plan Step Do Step Check Step Act Step Using the Problem Solving Tools and PDCA

Overview of the PDCA Problem Solving Cycle: The PDCA Wheel:

3 Overview of the PDCA Problem Solving Cycle: The PDCA Wheel PLAN DO CHECK ACT Design or revise business process components to improve results. Implement the plan and measure its performance. Assess the measurements and report the results to decision makers. Decide on changes needed to improve the process. The concept of the PDCA Cycle was originally developed by Walter Shewhart at Bell Laboratories in the 1930s. It was then popularized by W. Edwards Deming in the 1950s onward and is frequently referred to as “the Deming Wheel.”

Foundations of the PDCA Cycle :

4 Foundations of the PDCA Cycle The foundations of the PDCA cycle and Deming’s teachings consist of the following three principles: Customer Satisfaction: Satisfying customers’ needs should be paramount for all workers in the organization. Management by Fact. Decision making must be made on data collected from operations and analyzed using statistical tools. Decision makers must practice and encourage a scientific approach to problem solving. Respect for People. A sustainable problem solving and continuous improvement approach should be based on the belief that employees are self-motivated and are capable of coming up with effective and creative ideas.

Steps of PDCA: The Plan step :

5 Steps of PDCA: The Plan step Recognize the problem and establish priorities. Problem may be outlined in very general terms based on information from several sources. Form the problem solving team. Interdisciplinary teams of individuals close to the problem are best. Define the problem and its scope clearly. Who,What,Where and When. Pareto Analysis can be useful in defining the problem. Analyze the problem/process. Process flowcharts can be useful a useful tool. Determine possible causes. Cause-and-effect diagrams are helpful in identifying root causes of a problem. Data from the diagrams can be organized using check sheets, scatter diagrams, histograms, and run charts. Identify possible solutions. Brainstorm to find solutions. Avoid the temptation to propose quick, immediate fixes. Goals should be specific, measurable, achievable, and realistic. Evaluate potential solutions. Focus on solutions that address root causes and prevention of problem occurrence. Solutions should be cost-effective; achieving group consensus is important.

Steps of PDCA (continued):

6 Steps of PDCA (continued) The Do step Implement the solution or process change Monitor results and collect data The Check step Review and evaluate the result of the change Measure progress against milestones Check for any unforeseen consequences The Act step If successful, Standardize process changes Communicate to all involved Provide training in new methods

Problem Solving Tools:

7 Problem Solving Tools Check Sheet. A simple tool for collecting data about problems or complaints. Example 1.

Problem Solving Tools (continued):

8 Problem Solving Tools (continued) 2. Histogram . A graph which presents the collected data as a frequency distribution in bar-chart form. Example 1

Problem Solving Tools (continued):

9 Problem Solving Tools (continued) Pareto Chart. Orders problems by their relative frequency in decreasing order. Focus and priority should be given to problems that offer the largest potential improvement. Example 1

Problem Solving Tools (continued):

10 Problem Solving Tools (continued) 4. Scatter Diagram. A graphical tool to check if two relationships exist between two variables. Example 1

Problem Solving Tools (continued):

11 Problem Solving Tools (continued) 5. Flowchart. A visual representation of a process which can help in identifying points where failures may occur and intervention is useful. Example 2 Rounded box - represents an event which occurs automatically, such as receive telephone call. Rectangle or box - represents an event which is controlled within the process. A step or action which is taken. Diamond - represents a decision point in the process. Typically will require a `yes' or `no' response. Circle - represents a point at which the flowchart connects with another process. The name of the other process appears within the symbol.

Problem Solving Tools (continued):

12 Problem Solving Tools (continued) 6. Cause-and-effect diagram (fishbone diagram). Offers a structured approach for identifying all possible causes of a problem. The classic diagram is as shown: In retail, a better representation is the 5S (Pal & Byron 2003):

Problem Solving Tools (continued):

13 Problem Solving Tools (continued) 7. Control Chart. Is a statistical tool used to monitor the performance of a process over time. It is a time-ordered graph of sample data which can be used to identify when assignable causes of variation may be present. Example 1 If observations fall between the lower control limit and the upper control limit, then the process is in control. Otherwise, corrective action is needed. Upper Control Limit Lower Control Limit

Problem Solving Tools (continued):

14 Problem Solving Tools (continued) Process variations can be one of two types: Random variations which are created by many minor factors. Assignable variations whose main source can be identified and corrected. There are four types of control charts: 1. Control Charts for Variables. Variables are measured such as length of time a certain item is out of stock. Control charts can be: 1.a. Mean Control Charts 1.b. Range Control Charts 2. Control Charts for Attributes. Attributes are counted such as number of service calls and number of returns on an item. Control charts can be: 2.a. p -charts 2.b. c -Charts

Problem Solving Tools (continued):

15 Problem Solving Tools (continued) Mean Control Charts: Used to measure the mean (central tendency) of a process. Example 3: A store manager is concerned about the length of time items are out of stock. The store manager has a policy that the average time items can be out of stock is 2.2 days (30.8 hours based on 14-hour days) with a standard deviation of 5 hours. A sample of 10 out of stock items was taken and the length of time they were out of stock was measured. The data is shown below: LCL=Mean-2(Standard Deviation) = 30.8 – 2 (5)= 20.8 hrs UCL=Mean+2(Standard Deviation) = 30.8 + 2 (5)= 40.8 hrs Since 41.2 is larger than 40.8, the duration of out of stock has a mean larger than 30.8 hours.

Problem Solving Tools (continued):

16 Problem Solving Tools (continued) Range Control Charts: Used to measure the variability (dispersion) of a process. Example 3: Suppose that the average range for the mean stock out duration over the last 5 samples has been R 1 =32 hours Range = R = 61 – 18 = 43 hours LCL R = D 3 R =0.22 × 32 = 7.04 hours UCL R = D 4 R = 1.78 × 32 = 56.96 hours Since 43 is between 7.04 and 56.96, the dispersion of the stock out times has not changed.

Problem Solving Tools (continued):

17 Problem Solving Tools (continued) p -charts: Used to monitor the percentage of failures generated by a process. Example 1: Suppose that managers are concerned about increased number of complaints in Oct-Dec. Has the installation process deteriorated over the last three months? Using Jan-Sep data Proportion of complaints =103/9740=0.01057 For sample size of 100 installations for Oct, Nov, and Dec, Standard deviation of p = ((0.015057)(1-0.01057)/400) 1/2 =0.005113 LCL=0.01057-2(0.005113)=0.000348 UCL=0.01057+2(0.005113)=0.02080 Since all sample proportions are between 0.00348 and 0.02080, the process is in control.

Problem Solving Tools (continued):

18 Problem Solving Tools (continued) c -Charts: Used to monitor the number of defects per unit. Example 4: Number of defectives per 25 yd 2 of installed bathroom tile in 10 samples of 25 yd 2 are: LCL=Average–2 (average) 1/2 =1.90–2 (1.90) 1/2 = -0.857 LCL=0 UCL=Average+2(average) 1/2 =1.90+2 (1.90) 1/2 = 4.65 A new sample of 25 yd 2 was examined and the number of defects was 5. Since 5>4.65, the process needs to be examined.

Problem Solving Tools (continued):

19 Problem Solving Tools (continued) 8. Run Chart. A graphical tool to monitor important process variables over time. A helpful tool in identifying trends and cycles. Example 1 9. Brainstorming. An idea generating technique for a group of people to solve problems. Encourage unrestrained collective thinking All ideas are welcomed No single member dominates the session

Problem Solving Tools (continued):

20 Problem Solving Tools (continued) 10. Interviewing. A technique for identifying problems and collecting information. It can be used with employees for internal problems and customers for external problems. 11. Benchmarking. A process of measuring performance against best in the industry or the best in other industries. Steps: Identify the process that needs improvement Find the store that excels in the process Obtain data on the way they perform the process Analyze the data Implement process improvements in your own process 12. Fail Safe Methods. The implementation of methods and warnings to decrease probability of mistakes.

authorStream Live Help