Specs of the product are essential in in the Design Specification Document to ensure that testing and development can be properly executed. The specs of the product must be properly documented to allow for product development in the future. In addition, any product will have a range of error in development accounting for slight variance in size and shape. This must be accounted for prior, therefore when it is sent out to be tested on, the results can be verified. As for determining the specs, that largely depends on the product/device you are developing. In terms of the acceptable range of variance accounted for on your documentation, it is required that all means of failure and error be accounted for. For example, to submit the specs for testing and production of a device that will be 3D printed on a Mark Forged printer, it could be necessary to account for a dimensional inaccuracy of up to 125 microns.

To start off, the specs that should be specified for a medical device should only be directly related to its functionality, as well as interaction with the user. Functions such as image accuracy and scan speed of a dental scanner, for instance, would be appropriate specs to be specified because they directly define the functionality of the product. In addition, the scanner's weight, size, and dimensions are some of the specs that correlates with its usage by the user. Now, the extent of those specs, primarily the functionality specs, must be carefully determined by thorough testing conducted by the company. Companies can only advertise their medical device with the specs that the device can safely and effectively reach. Therefore, determining the true potential of a medical device is crucial and necessary for determining the extent of its specs.

To me, this is probably one of the hardest part of the design process. There are many problems associated with creating tolerances for various aspects of the product. On the one hand if you make the tolerance too tight, it may be difficult for manufacturability or become extremely expensive when ordering from your vendors. When the tolerance is too loose, you get parts which don't assemble right or features which do not work. For example, if you have a hole which is on the lower control limit and a pin which has a diameter on the upper control limit. This combination may make it so the pin does not end up fitting into the hole and there is a failure. This is how tolerance stacking works and can be a difficult calculation sometimes. Sometimes, it all just comes down to trial and error and running endless combinations to see which dimensions for certain features will result in a successful product.

Specs is determined during the test of the product accordingly to that the specs is also upgraded. As the product is fabricated there will be chances of some specs getting destroy in the manufacturing site. So that the reason product goes into multiple tests which comes out as final and better product.

I mostly agree with what what said above. I believe there is a hierarchy of importance. First, the device needs to be within tolerance of it's designed use. You can't have an implant that is too big or small from it's designed size. Next, you need to have the correct tolerances in the parts. If the pieces won't fit together properly, it's not a working product. Lastly, the tolerances allow for machining differences. If you required exact changes, you could make a few in spec, but soon after due to natural changes/differences in the machinery you would fall out of spec very quickly.

From my experience working on projects, deciding on detailed specifications may seem to be the most daunting task, but becomes clear-cut the more you are familiar with the process. The most important factor when specifying ranges of specs is the limiting conditions involved in the design input. I work on designs for a Solar Car team, and when we decide on design specifications, the ranges for design inputs are either based on earlier inputs that we set as a condition to meet or design regulations. For example, FSGP regulations requires that the vehicle must have two inches of ground clearance, so this sets the lower range for many design specs that will go into the suspension systems of the solar car. Sometimes it is based on the objective of what we want to achieve with a next generation solar car. Say, we want straight-line efficiency to take precedence over cornering efficiency in the next generation solar car, then the tolerance for specs that directly affect the rolling resistance (leaning of a car while cornering) can have a wider range without significant concern. Another factor that helps decide on a range is simply following standard equations that are used in specific designs that can help allude to desirable ranges of a design input. There are several design limitations that are already in place that helps deciding on detail specs a lot easier.

In my experience, a lot of specifications have a tolerance enough where the manufacturer is able to make the product well and it doesn't compromise the integrity or functionality of the product. In certain cases, there have been instances in which the manufacturer physically cannot make the range due to their equipment. In these instances, the engineers between both companies would meet and discuss what is possible and what is not. If the conclusion from this meeting is that the manufacturer cannot make this product, then the company needs to outsource to another company that can make the product as per the specification. I've seen that in most cases that the engineers between both companies can come together and make a viable plan to adjust the specification. This issue is rare as the engineers of one company keep in mind to keep the tolerances to a level where the manufacturer can make the product without tedious issues that can hinder production.

I agree with what has been said above, that determining specs is an extremely important and delicate part of the design process. The specs must be tight enough so that the product can provide its intended function and also loose enough so it can be mass produced by machines with low failure levels. This can be a delicate balance to achieve, especially when dealing with machines that mass produced product. Extensive testing is necessary to determine the correct specs, as well as working closely with manufacturers to ensure the specs are doable. 

Determining the spec with tolerances is the most important part in the effectiveness of the design. It really depends on the application and market in my opinion. If it is an orthopedic implant, there should be a plan included to design, develop, and validate multiple sizes of the same implant to accommodate patients with varying size of the body part. The initial size should be an average size for a subset of people, from data taken from anthropomorphic data or other accurate sources providing average human limb sizes. 

The first time setting specs and tolerances are usually wrong. The real challenge is learning from failed validation results, and how to change a design that will improve validation results. 

Many successful projects might take so long because of the trial and error process, but ultimately gets a better design in the long run. 

@amm7 I agree that finding the right balance between tight and loos specification is very important however I would also like to note that beyond functionality is the impact on long term performance and regulatory compliance. What do I mean by this? A spec that is too loose might initially pass but lead to premature failure in real world use. Similarly, testing should not only focus on the ease of production but also on environmental factors, degradation, and even other stress conditions. Collaboration with the manufacturers is crucial, but this does not mean to overlook the input that material scientists and reliability engineers can offer. This ensures that specifications are not only practical, but also durable.

I think the starting point has to be a deep understanding of the user needs, regulatory requirements, and the intended use of the device. From there, we can also pull information from previous product data, industry standards, feasibility studies, and early prototype testing to figure out a realistic and safe performance range. It's important to set specs that are achievable but still tight enough to ensure product quality and consistency. As for how broad the spec can be — that's a balancing act. If the spec is too narrow, you risk a lot of unnecessary failures and rework during testing. But if it's too broad, you might end up with a device that doesn't perform reliably or could even pose safety risks. I think the key is to make sure that any expansion of the spec is backed by strong risk analysis and clinical or engineering justification. If expanding the range introduces any new risks or compromises the core function or safety of the device, that’s when it's gone too far.

To determine an appropriate specification range, I would start by grounding it in user needs, intended use, and real-world operating conditions, then translate those into measurable design inputs supported by prior data, standards, and risk analysis. Historical performance data, pilot testing, material capabilities, and relevant regulatory or industry standards can help define realistic upper and lower bounds that the product can consistently meet. Verification and tolerance analysis are key to ensuring the spec reflects true product capability rather than ideal performance. A specification becomes too wide when it no longer meaningfully controls risk or ensures performance, such that passing the spec does not guarantee the product meets user needs or safety requirements. Expanding a spec should only be justified if risk remains controlled and performance is not compromised, and any expansion should be formally documented through design controls and change management to maintain traceability and regulatory compliance.

In a scenario where it was required of me work on a product and I needed to determine an initial range for the product grow, I would first go into the particulars of the product to truly understand what heights the product can reach. For instance if this was a drug infused medical device, it would be necessary to understand whether the device has utility beyond what was originally established by using on aspects of a person's body. So determining the product's spec requires directly addressing a clinical problem by keeping the research and testing to topics that specifically pertains to that clinical problem. This acts as a way to not only ensure that the product remains focused, but as a way to offer more legitimacy towards the product which will cause the possibility of the scope to expand. Delving into other topics and expanding the spec of the product should only come about through thorough testing and classification of the original route cause of the problem that the product addresses. 

It is extremely important to have justified specifications for each design input when following design controls for medical devices. To determine specification, you would look at each design input individually and define a measurable value or range that meets that user need. These values can come from peer reviewed literature, historical data from similar products, or early prototype testing. The tolerance of a specification should be based on performance, safety, and ability to manufacture the device within that range. It should only be expanded when testing or historical data shows that these are still met within an expanded range. From my experience defining design specifications for a project in undergrad, I mostly used data from peer reviewed literature. The device I was developing was a training model of the arm used to teach physicians how to perform ligament integrity tests and required that the materials used to represent soft tissue have similar mechanical properties to human tissue. For each specification related to mechanical properties, my team used the mean and standard deviation of human tissue to determine the specification range. This is just one example of how specifications can be defined to ensure they are justified and defensible under design controls.