**Problem definition.**

In the previous blog posting an analysis on the learning curve over the blog posting was conducted. The main target of the post was to forecast the production time for each blog posting during the upcoming weeks of the program. This will then be used to re-baseline the blog posting project. However, before initiating this process the data of the blog posting should be first validated and analyzed. (SPC) Statistical Process Control method will be used to analyze the data in order to determine any variation in the process and consequently control it or perhaps even optimize it.

Therefore the topic of this blog will be analyzing statistically the process of blog posting using SPC methods and explore manners of improvements.

**Identify the feasible alternative.**

Statistical Process Control (SPC) is a method for monitoring, controlling and improving a process through statistical analysis. It’s a tool used in the industry to measure and control quality during the progress of a process (more specifically manufacturing process). It analyzes and determines the frequency of variations in a process and hence used to optimize the process by reducing these variations.

The quality data whether it were product or process measurements are taken during the live (on-line) manufacturing process. This data is then plotted on a graph with pre-determined control limits. Control limits are determined by the capability of the process, whereas specification limits are determined by the client’s needs. [2].

To implement SPC the following steps should be taken [1]:

- Identifying the measurement method
- Qualifying the measurement system
- Collect the data and initiate the SPC Charting
- Reaction plan development and documentation
- Include chart to the control plan
- Calculate the control limits
- Assess the control process
- Analyze the data to identify root cause and correct
- Design and implement actions to improve process capability
- Calculate CP and CPK and compare to Benchmark
- Monitor and focus efforts on next highest priority.

The following image illustrates the process to undertake an SPC analysis on a process

For the purpose of this paper control charts will be used to analyze the SPC data from the blog posting project. This is mainly done to detect any variation on the process of the blog posting development as well as controlling the process and probably improve it if required.

It should be noted that there are many types of control charts that is being used in the industry. The type of charts depends on the type of the data as following:

- Variable data: measured and plotted on a continuous scale this includes time, temperature cost and figures
- Attribute data : measured on discrete events such as shipping errors and percentage of waste

**Selection of Criteria**

As the data collected and measured for the blog posting is time the type of control chart to be used is Average and range chart with sample sizes <10.

**4. Development of the outcome for alternative.**

As per the previous blog the following data on the blog posting were collected:

Weekly Blog Posting |
||||||

Weeks | 1 | 2 | 3 | 4 | 5 | 6 |

Time Hrs. | 3 | 3 | 3 | 2 | 2 | 3 |

For a variable data such as the described the following data were developed to produce a control chart with± 3x Deviation.

The above data was then translated to the following chart:

Limitation: the data collected for the blog posting is relatively for an analysis such as the above. Usually the requirement is to collect around 20-25 samples before calculating the statists and control limits. However, since this is a validation exercise this tool would suffice to analyze the blog posting

** ****5. Analysis of the alternatives:**

The data shown in the above chart illustrates a controlled process. As all the points fall within the boundaries of the control limits as well as it being spread around the mean line. Though, it could be noticed some slight fluctuations of the plotted points within the control limits which could be an indication of a change in the process. The change in process for the blog posting project, could be interpreted as the learning time required for any new tool used in each blog post. The process seems to be in-control.

Another point could be detected from the above chart is the location of the Central Line (Mean) which sits around 2.6 Hrs. As per the specification of the project it is more than the estimated time required for the blog posting in the original plan (which was 1 hour).

**6. Selection of the preferred alternatives**

**T**he conclusion of the above analysis indicates that the average time required for the blog posting is around 2.6 Hrs. This proves that, the exercise conducted in the previous blog post analysis is valid as the proposed range was between 2.1-3.00 hours.

Therefore for the re-baseline exercise the chosen time for the blog posting will be around 2.6 hours.

** 7. ****Performance Monitoring and the Post Evaluation of Result.**

To maintain the status of the blog posting process as being in-control, a continuous updating to the SPC of this project to be conducted. Any deviation from the process will be illustrated as out of the control limits. This would warrant a reaction to tackle the change and improve the process.

** **

8. **References**

- 1. More Steam (2014). Statistical Process Control (SPC). [ONLINE] Available at: https://www.moresteam.com/toolbox/statistical-process-control-spc.cfm. [Last Accessed July 14, 2014].
- 2. InfinityQS (2014). What is Statistical Process Control (SPC)?. [ONLINE] Available at: http://www.infinityqs.com/resources/what-is-spc. [Last Accessed July 14, 2014].
- 3. Edgell R (2014). Statistical Process Control Explained. [ONLINE] Available at: http://www.winspc.com/what-is-spc/statistical-process-control-explained. [Last Accessed July 14, 2014].
- 4. Brassard M, Ritter D, (2010). Memory Jogger 2. 2nd ed. USA: GOAL/QPC.

AWESOME posting, Ms. Mahfoodah!! Unfortunately, you don’t have many readings yet but at least so far, the process appears to be “in control” and “capable”……

Just be sure to turn to page 63 in your memory jogger. In addition to OUTLIERS (which you don’t have yet but I bet if you plotted EVERYONE’s hours and not just yours, you would find some) but also you need to watch for PATTERNS. (ref page 63)

If you have an OUTLIER that is known as a “special” or “identifiable” cause that is normally caused by an EXTERNAL factor not related to the process itself. IF you have an outlier, you do NOT include that data point in your analysis…

The more worrisome kinds of problems are the patterns, which indicate something is wrong INTERNALLY within the PROCESS. (The questions you will likely see on your PMP Exam are 2D or 2C but there are others)

Anyway IF you can get the data, for your W7 blog, why not try to get all the data from everyone’s hours (it should be available from the time sheets?) and plot EVERYONE’s time and see what you come up with? When you create your W7 Statistical Process Control Chart, be sure to put in not only the upper and lower control limits (+/-3 sigma) but also the ZONES- +/-1 Sigma and +/-2 Sigma.

Bottom line- AWESOME case study and you did very well on it, but let’s do another blog so you can explore this very important topic in more detail.

BR,

Dr. PDG, Jakarta