For predictive measures, I don't think a team would decide to log more inputs than necessary. The intentions of the project might be to go above and beyond however at the end of the day it's a business and you'd want to give the optimal output for the customer with the maximum return for yourself, and logging more inputs might cost the company more. If they were to reach out to the customer and discuss with them about providing a more exceeding product then yes of course the project can advance past those user needs. However, a team can definitely log more inputs than necessary for unpredictive measures. For every cause there's an effect and of course, the engineer would be the one to find mishaps when trying to meet the customer's needs. So additional inputs might need to be set to reach the desired output. I also believe that "over spec-ing" is a thing and there should be a fine balance between your validation and verification or in other words a balance between your engineering desires and customer requirements to avoid complications.

Typically, several inputs are determined based on the customer needs provided by marketing and the VOC evaluation done by R&D. These inputs are then consolidated to a list that is both extensive enough that it covers all needs but also not superfluous and repetitive. Most companies are very cost conscious and teh design team plans with validations in mind.

Companies may take the wrong approach to metrics if they measure only results obtained over a long period of time (retroactive). An alternative approach is to supplement the standard performance metrics with predictive metrics. This type of metric measures is a driver of behavior by being more process-oriented than performance-oriented. Predictive metrics are leading indicators that show you are creating the outcome you want to create, day-by-day, and month-to-month. Measurements taken during the course of a longer process show the direction in which a continuous improvement initiative is headed. I guess excessive inputs can be hazardous especially if they are insignificant or irrelevant. It is definitely possible for a project to advance past user needs in hopes of delivering a better product than intended. It would probably be beneficial for the company's future to have this.

https://kanbanize.com/blog/predictive-metrics/

Kim, J. O. (1971). Predictive measures of ordinal association. American Journal of Sociology, 76(5), 891-907.

Although I do not have industry experience I would say I was taught a good lesson by a professor in my senior capstone project. After getting our project and our customer needs, one of my group mates had many great ideas that were additional inputs to our already decently long list of customer needs. After my group-mate told his ideas to our capstone advisor, our advisor gave us this good piece of advice, "It's better to under-promise and over-deliver than over-promise and under-deliver". Obviously a project should always strive to make a better project than asked of, but not at the cost of underwhelming the customer. I think that a project should make the given inputs the main priority and try to limit additional inputs. Additional inputs can cause delays in the project's stages and potential other issues for something that was not necessary. Therefore, projects should rather focus on meeting given inputs with quality rather than have to focus on additional, non-provided inputs. 

I do believe that over-validation can easily occur if the proper protocols aren't followed during the initial and validation phases of the design control process. If we look at most of the medical devices that have been along for a few decades, they all have some small flaws that may have prevented the device from being released if over-validation had occurred. As mentioned in the lectures, we would also have to think about all the documents in the process that would need updating/new revisions if additional outputs are created due to over-validation. I think overtime, all departments involved in the planning phase of design controls for a company gain sufficient experience on how to avoid over-validation by creating ample outputs that would meet all customer needs and ensure safety of the product.  

As mentioned in the lecture and by others on this post, when determining inputs for a device the marketing team is included in the discussions. The customer needs research done by the marketing team would ensure that the device inputs are necessary and wanted. Ideally, this would allow inputs that are not needed or wanted from being included. As for companies including additional inputs to make a better product, there will be time and financial hurdles to prevent unnecessary or unneeded inputs from being included. Furthermore, including too many inputs or features in a product may make it unappealing or overly complicated to consumers. 

@devdesai I am glad you brought this up because honestly this is one of the most interesting things I have learned in this class so far. I had no idea how important the marketing department was to the development of a medical device. I had assumed (incorrectly) that marketing focused of making marketing materials to help the sales team once a product was already developed and ready to be sold. I agree that including the marketing team during this part of the development phase is vital because they have direct contact with every group relevant to the production of a device so their knowledge base is larger that the engineers and so they are able to steer development in the right direction to keep them on track.

I 1000% believe this occurs more often than we all think in every industry. At my previous job, I worked a lot with a certain ancient telecommunications company. As a contractor, we had access to their project trackers and other planning documents for our own planning purposes. Simple projects that shouldn't have taken more than three months from beginning to end had been aging for years. Projects started in 2020 haven't been completed until today. It's not like the work we were doing was overly complicated; this company refused to modernize and streamline their equipment. Everything was bloated with unnecessary checks, money wasted, and time squandered. So yes, it is very possible to over design from the top-down, this is why I assume Dr. Simon preaches including the "ground troops" of the company in the planning phases of projects.

I think over validation may occur especially if your boss is too neurotic. Project planning is that's why very important as at that time you have to plan the timeline too. If you add too many features in a short time it may lead to past due date. 

When a company is preparing the project definition form, the market department should already conclude the requirement of the customer and engineering departments. The project's input is supposed to be the same as the customer's need. If a team decides to make additional input to the project. That will increase the cost of this project. Finishing the project and the product fulfills the validation in time is the most important thing for a company. It's possible to add external ideas or modifications to the design anytime, but the company's procedure to run design change requests for a project might be different.

I also do not have any industry experience in design work, however, when I completed my senior design project the customer ranked their inputs in what was necessary to what was not as important. When we were designing the project, we had to accommodate what they said were non-negotiables and we did our best to incorporate the other aspects that they would have liked. If we had to incorporate every single input, they had we would not have been able to finish the project on time, nor would it have been on budget.

In industry is this a similar concept? The more inputs and validation/verification that is required the more money and time product will need. Is there a point where they draw the line, and you must pick the most important ones that will still meet the important user inputs? I know from class that industry is more deadlines driven than other places, would that limit their ability to over verify products?

I also do not have any industry experience in design work, however, when I completed my senior design project the customer ranked their inputs in what was necessary to what was not as important. When we were designing the project, we had to accommodate what they said were non-negotiables and we did our best to incorporate the other aspects that they would have liked. If we had to incorporate every single input, they had we would not have been able to finish the project on time, nor would it have been on budget.

In industry is this a similar concept? The more inputs and validation/verification that is required the more money and time product will need. Is there a point where they draw the line, and you must pick the most important ones that will still meet the important user inputs? I know from class that industry is more deadlines driven than other places, would that limit their ability to over verify products?

I don't truly feel like there could really be design over validation since you would end up with a limited time frame set by your company to get the device into pre-clinical and clinical trials. In my opinion, it is best to try and provide many different types of evidence and providing a lot of data to prove that the device is efficient and safe, especially if you are trying to get the device approved to go inside the human body. If the FDA feels that not enough evidence is provided, they will ask to provide more data and evidence to make sure that the device is sufficient and is actually providing a reliable and effective treatment. In my job, I have seen this happen where clients would have to come back and ask us to process a few more runs because the FDA asked for more supporting data. I do think that when designing a device and picking what tests that need to be performed, that you need to decide which tests are the most important to perform for your device and which can just be supported using research that has already been done. 

In my opinion, the company will avoid it as overvalidation will be a cost to the company. If you provide more features than necessary, this will incur additional costs to the company and the product to be produced at each stage. this may be seen as good by the end user, but as the cost will increase, the money that the end user will pay to buy the product will also increase. For this reason, it should be well thought about while determining the features for validation, extra inputs that will create cost and that are not necessary should be avoided.