Hi! I agree with all of the previous posts that when an experimental failure occurs, it’s important to define what went wrong and perform further testing. However, I want to add that to be successful in performing these tasks following an experimental failure, lab members must remain calm and understand that experimental failures can always happen, even with the most experienced lab member. Although failure is not desired, it can happen, and solutions exist. Remaining calm when failure occurs is not a single action that needs to be taken to resolve the issue, but it is the first step that needs to be taken to ensure that other appropriate actions are taken. For example, during one of my experiments, there was an issue with data collection where one of the required variables was not collected properly and I started panicking. Although I knew exactly what the problem was, I couldn’t think of a clear solution until I stopped panicking and thought of the problem clearly, and in the end, everything worked out. To sum everything up, stay calm and refresh your mind when you run into an experimental failure. Don’t panic and think everything through clearly to be able to identify problems and solutions.
I understand that some failures are much more severe than others. Are there any experimental failures that have no solution? Have any of you experienced any of these failures? I’m interested in reading more about this!
First, we need to spot the error. What caused the error, and what condition affected the test? These give us better clarity on how to tackle it. Re-read the procedures, and understand the specification it is been asked. We need to understand the mistakes so we can improve from that. More testing is definitely a go, but implementing different variables or conditions could be ideal. Maybe the materials we are using are not of great quality, so we might need to change the supplier. We need to meet the specific guidelines, all we have to do is modified how the test is done, however, we cannot alter the ultimate intention of the product we are working on.
A good rule for any experiment is that there is no such thing as bad data, data is data. So when something fails or goes wrong in a project, first thing is to record what happened. This information can be used as an example or referenced later on and can serve as a learning experience. Next is to investigate what went wrong. In a more larger, organizational scale this can be determining root cause or root cause analysis. Once we know the cause we can adjust our procedure and/or parameters in the project to get a desirable result. Like with the simulation we are figuring out why the label did not work and what can be done with the testing to alleviate this issue.
The most effective method for resolution is to simply retest. Of course before retesting it is important to analyze the results to get a better understanding of where exactly the test failed. For example in simulation I the only area that did not pass were the peeling of the labels. Knowing this information our group focused on how can we improve the quality of the adhesion of the labels rather than worrying about other specifications that already passed.
The most effective method for resolution first starts with clearly identifying the main issue with the experiment. In industry, specifically in quality engineering, this is most commonly done through a process called root cause analysis, which involves analyzing the process through cause and effect to conclude the main issue. The cause and effect can be observed by utilizing the 5 whys, which consist of asking five questions until the main reason is revealed. The purpose of doing this is to ensure that the problem is solved with a more permanent solution to ensure that it is not a temporary fix. Upon finding the root cause, a solution can be designed to ensure that the experiment is resolved. In the medical device industry, most companies may utilize a Corrective and Preventive Action (CAPA) in order to ensure that the process is fixed going forward. A CAPA typically consists of different steps such as action planning, implementation, and effectiveness verification, in which the solutions for the issue are mapped out and tested, before becoming a permanent change.
When I was looking at the simulation instructions, there were several ideas popped up in my mind. The first thing I suspect is, does the employee follow the instruction strictly? Then I will suspect that the sticker and bottle vendors provided are defective products. The verification step is the last thing I examine. I didn't work in the industry before, so when I discussed it with my team members, I realized that my thought was quite different from others. I am used to examining the instruments I used first to figure out where the problem is in the experiment. I believe people would do the instrument test in the industry, but I want to know what the order of the instrument examination is in a company.
I believe problem-solving and revision is the best way to resolve the problem. While reading through the verification test, my group noticed that several things were missing that were deemed important in the DID and DSD. We also saw that the procedure in the testing was performed incorrectly as well. Through the process of elimination and figuring out if the inputs equated to the outputs, we were efficient in finding a solution.
I believe when we are dealing with such a situation, it's important to follow a systematic and structured approach to resolve the issue and determine its root cause. Documenting the failure can be the first step; a detailed description of what went wrong, when it happened, and any relevant data that might help the investigation. This can be followed by gathering and collecting data and evidence related to the failure, such as log files, error messages, performance metrics, etc. Then, we need to analyze the data meaning that use the data collected to identify the root cause of the failure. This may involve reviewing code, comparing performance metrics, or reviewing system logs. Then developing/designing a resolution plan based on the root cause, and developing a plan to resolve the failure is needed. This may involve fixing a bug, reoptimizing or making a change to the experimental setup. After implementation of the resolution, we need to test the resolution to ensure that it has effectively resolved the issue and does not cause any unintended consequences. Last but not least is documenting the resolution as it may happen several times. This protocol should contain the steps taken to resolve the issue, the resolution itself, and any lessons learned for future reference. I think by following these steps, we can ensure that experiment failure is resolved in a systematic and effective manner and that the root cause of the issue is accurately identified and addressed.
When dealing with failures during testing, I believe that reviewing what went wrong, altering the testing procedure and keeping a level head is extremely important. Keeping a level head is extremely important in the field especially when failures happen because the failure could always be fixed. Not letting the failure cause you to shake the confidence in your skills and bouncing back to try and resolve the issue is a very big quality that employers tend to look for in potential candidates. By keeping a level head, everything else will fall into place such as there was an extra step needed in the procedure or altering how the product is testing. I think the bottle simulation is extremely important in showing how a solution will be found eventually as long as you are open to any possible solutions and suggestions made and be willing to look at the failure as a chance to expand your knowledge.
The most important part of dealing with experiment failure and coming up with a resolution is to be able to go through the experimental process thoroughly to understand why the experiment was performed, how it was performed, and what caused certain outcomes. As my team was working on this simulation, we went through every possible solution such as, changing label vendors, making them waterproof, adding friction between the bottle and the label, changing the bottle composition, etc., and all these solutions only seemed to complicate the product or complicate the process of problem resolution. Because all of these would’ve required major changes to the product, like finding new vendors for bottles/labels, or making sure that the new bottle composition wouldn’t interfere with the AquaWoah solution. But when we went back to the experiment or the verification test and understood why and how the test was performed a certain way — the solution came easy to us. We just had to change how the verification test was done, not the product itself. This is an example of why it is important to understand the problem you have and what you are trying to solve.
The most important part of dealing with experiment failure and coming up with a resolution is to be able to go through the experimental process thoroughly to understand why the experiment was performed, how it was performed, and what caused certain outcomes. As my team was working on this simulation, we went through every possible solution such as, changing label vendors, making them waterproof, adding friction between the bottle and the label, changing the bottle composition, etc., and all these solutions only seemed to complicate the product or complicate the process of problem resolution. Because all of these would’ve required major changes to the product, like finding new vendors for bottles/labels, or making sure that the new bottle composition wouldn’t interfere with the AquaWoah solution. But when we went back to the experiment or the verification test and understood why and how the test was performed a certain way — the solution came easy to us. We just had to change how the verification test was done, not the product itself. This is an example of why it is important to understand the problem you have and what you are trying to solve.
I fully agree with this post here. Knowing the ins and outs of the testing process is paramount to determining the cause of failure. As such, I'd like to suggest that documentation is the most important factor of testing. If all the tests are not designed well, the results are skewed, and there are no control specimens, but all the minutiae of the process are diligently and accurately documented; the invalid processes and results can be used to adjust the testing process. Keeping records is the most important part of good testing.
In the simulation this week, we were tasked with correcting bottle label failure. When dealing with experiment failure in general, what do you feel is the most effective method for resolution?
Dealing with project failure is an important part of being a project leader. There are many ways to determine what went wrong during an experiment. One effective approach is to conduct a post-mortem analysis with the team to identify the root cause of the failure and develop strategies to prevent it from happening again in future projects. Another way is to seek feedback from stakeholders and incorporate their suggestions into the project plan for better outcomes. Firstly, it may be helpful to thoroughly examine the goal of your experience, if you have not already done so. This can help you understand why the experiment failed. Additionally, it would be useful to break up your experience into multiple parts, examining each section to see where it could go wrong. By doing this, you can identify areas that need improvement and make the necessary adjustments. Furthermore, it is essential to communicate effectively with group members to ensure that their feedback is also understood and implemented appropriately.
