Continuous improvement is a basic element of modern industry and there are many tools and methods available to help with continuous improvement. Some situations, such as reducing manufacturing costs by 10%, may require the use of the Six Sigma methodology. For other situations, such as root cause analysis, the Plan-Do-Check-Act (PDCA) and basic quality tools may be sufficient.

In this example of a quality improvement project the performance of a catapult needed improvement due to a high degree of variation in the distance traveled by a projectile. Basic quality tools and PDCA were used to analyze the problem and implement a successful solution. The Plan phase consisted of investigating the problem using flowcharts and a fishbone diagram. The Do phase consisted of implementing the corrective actions and the effectiveness of the improvements were verified in the Check phase. The final phase of PDCA was where other opportunities for implementing the corrective actions would be sought. 


pdca flowchart
Figure 1: Flowchart

The first step after identifying the problem was to understand the catapult process. Quality Companion by Minitab was used to create a process flowchart, which is used to display the order of the steps in a process. Although the flowchart would be created on a computer; the real start of a flowchart is at the location of the process being flowcharted. In this case, in a basement hallway.

The catapult process was observed in action and notes were taken to use when creating the flowchart. The first step was the positioning of the catapult. Next, the rubber band was attached to the base and then attached to the catapult arm. Then projectile was placed into the cup on the catapult arm. The final operation was the firing of the catapult; this consisted of pulling down the catapult arm and then releasing it. The observations were entered into the flowchart, which can be seen in figure 1.   

Fishbone diagram

The flowchart was used to create an Ishikawa or fishbone diagram using the program Quality Companion as depicted in figure 2. According to Kaoru Ishikawa in Guide to Quality Control a fishbone diagram “is useful in sorting out the causes of dispersion and organizing mutual relationships.” The problem under investigation is the dispersion or variation in the shots fired so variation is listed at the front of the fishbone diagram. This fishbone diagram uses the categories man, machines, measurements, environment, methods and materials; however, other category names are an option. The next step was to identify the potential causes under each category. This was done by first observing the individual steps of the flowchart. Regardless of how well a process maybe understood, valuable data can be collected by direct observation. The process was operated several times and observed and only then did a brainstorming session take place to fill out the fishbone diagram. The factors and potential causes were then thoroughly investigated and the results are listed in table 1. 

fishbone chart man machines materials six sigma
Figure 2: Fishbone Diagram (click image for larger version)
Operators: There were two potential operators and it was discovered that they did not use the exact same method for manually releasing the catapult arm. 
Catapult: The catapult arm impacts the arm stopper and this may result in slight movement due to the force of the impact on the arm stopper and could move they entire catapult. 
Measuring device: The measuring device was checked to ensure that it was measuring correctly and the operator was correctly using it. No problems were found.
Projectile: The projectile had an uneven shape and some parts had a lower coefficient of friction than others, which could result in varying degrees of sliding after impact.
Catapult release: The catapult arm was released by hand and this could result in variation if the arm is not released in a consistent manner. 
Illumination: Certain parts of the test hallway were dim; however, the launch and landing areas were sufficiently illuminated. 
Floor: The floor was observed to have varying degrees of surface roughness. This could greatly influence variability due to the potential for the projectile to slide varying distances after each landing. 
Table 1: Fishbone Factors

Corrective actions

Corrective actions were taken based on the causes identified in the fishbone diagram. A method of releasing the catapult arm needed to be defined to ensure different operators release the catapult arm in a consistent matter. A work instruction would be an appropriate way to document the proper procedure. The arm stopper was secured into place to remove potential free play and the catapult base was secured to the ground to prevent potential movement after firing. The projectile was replaced with one that had the same weight; but was symmetrical and had a rougher outer surface to reduce sliding after impact. An electric release device was considered to ensure that the catapult arm is always released in a consistent matter; however, this option was considered too expensive to implement. The impact area was covered in a towel to present a rough, yet consistent surface to reduce variability. 


Quality improvements need to be verified after they are implemented so the last set of 10 shots prior to implementation of the improvements were checked and compared against 10 shots fired after the improvements were implemented. Minitab® Statistical Software was used to generate the boxplot depicted in figure 3, which shows the difference in variability before and after the improvements were in place. The improvements were a success and any future catapult will incorporate the improvements. Perhaps the next model will even have an electronic catapult arm release.     

pdca boxplot graph diagram
Figure 3: Boxplot

Lessons learned

It is important to ensure that the lessons learned during the quality improvement analysis are recorded so that the same mistakes will not be made in the future. This is especially important for potential improvements that were not implemented; such as the electronic catapult arm release. One particularly effective method of saving the information is to update the failure Modes and Effects Analysis (FMEA) for the part or process that was improved. 
Related parts and processes should also be reviewed to determine if the improvements should be implemented in the other parts or process. An effective solution should become standardized. The deciding factor should not be whether or not the problem has occurred elsewhere; rather, whether or not the problem could occur. 

The Benefits of PDCA and Quality Tools

Once cycle of PDCA may eliminate a quality problem; however, PDCA should be viewed as an ongoing cycle for quality improvement. The variability in the catapult has been reduced to an acceptable level, but is further improvement possible?  Additional analysis supported by PDCA could be performed to shift the process mean to the nominal of 700 cm. The impact area needed to be covered in a towel to prevent the projectile from sliding and this increases production costs due to the need for additional material; in this case the towel. Further investigation may identify a projectile design that reduces variability without the need of an additional towel. Making changes to a process has a risk of degrading process performance; using PDCA ensures that the results are effective.