This interdisciplinary course focuses on sustainable innovation, introducing entrepreneurial students to the realities of problem identification and solution design within the complex world of healthcare.
Concept Selection
Loading...
From the course by Duke University
Healthcare Innovation and Entrepreneurship
84 оценки
From the lesson
Designing an Innovation
What approaches work best for design creation? Is "Brainstorming" an actual process? How do you select the "best fit" design from competing choices? How do you assess the operating environment? Is regulation always bad? Do you have a reimbursement plan?
Meet the Instructors
Marilyn M. Lombardi, Ph.D.Associate Clinical Professor
School of Nursing
Bob Barnes, PhDInstructor
Biomedical Engineering
Ladies and gentlemen, this is Bob Barnes on behalf of Marilyn Lombardi.
Last time we met, we were talking about concepts and design, and
at this point you should be pretty well on your way to developing conceptual designs.
We're going to be talking about how, among the many designs you could come up with,
a way of selecting which one might be most efficient, most proficient, for your need.
As you know, we're working on designing an innovation.
Today, we're talking about concept selection.
In addition to everything else we've talked about, in terms of stakeholders and
the terms of their requirements, in terms of the need itself, in terms of
background, in terms of significance, in terms of competitors and/or alternatives.
There are six items that you need to be concerned about in your design,
one of which is the information for your marketing assessment.
As you begin your design,
you may indeed determine there's more market assessment you need to do.
The design of your product into a process consistent with
taking advantage of the market itself, the need you're trying to address,
the stakeholders, and to be able to do that within a competitive environment.
This freedom to operate which we're going to talk about a bit more after
this lecture, you will need a detailed plan of what
you're going to do after you finish your needs specification.
Essentially, your needs specification is going to take you to the point at which
you will be ready to move to the development of a prototype and
to begin to test that prototype and fine tune your design based upon
the reaction to the prototype or how the prototype is experienced.
You'll need a project team to do this and
darn it, you're going to need some cash or some capital to make this happen too.
What will happen is, depending on the amount of capital that you can raise,
the market that you have, you will keep looking at all six of these
factors over and over and each one will have a tendency to constrain the others.
The idea for this little picture is if you change any one of these six attributes or
these six constraints, the others will be affected,
whether it's your product, or your process, your plan etc.
As an entrepreneur innovator,
you have to always be thinking about how do I make these changes and
how do I make these changes within the time frame that I have to get to market?
Let's take a look at a requirement and a need based upon
work that some of my students and some of my clients have done in the past.
We're going to look specifically at an operating room environment.
This particular team identified about 25 items, or 25 rough cut need statements and
they reduced those down to six that they felt were
worthwhile to investigate in some detail.
When they did that, they came up with some selection criteria and
you'll see some of these are weighted zero.
What that means is when they started looking at these selection criteria
carefully, their relative scores were so low,
they decided just to leave those out completely.
But I thought I'd show all of those to you so you get an idea, or
a reminder of how the selection is done and the selection criteria is developed.
As a consequence of the weighting and the scoring of these six,
they chose as the need that they wanted to spend the most time with is
there's a need to reduce chord clutter in the operating room.
Obviously, there two others there that are very close to that in terms of scores, but
in terms of what this team was very interested in,
they said this is the need we want to work with.
Based upon the constraints that we face in terms of marketing and
our freedom to operate the team members that we have and their skill sets and
what we think would require in terms of capital.
Now once they had decided what their need was,
they took a look at the stakeholder requirements.
The requirements, and you can see these, these cords out of the surgeon's way,
the nurse's path.
We started looking at this specific need and then asked if we
addressed that need, what requirements, from a stakeholder's perspective,
must whatever we develop address in order to be acceptable?
At this point,
we have a need that we've chosen based upon a certain set of criteria.
Now, we've looked at that specific need and
looked at the stakeholders a bit closer.
Now, we said based on our understanding of the stakeholders, this is what we think
the stakeholders want, whatever product we develop to be able to achieve.
This team came up with a number of alternative concepts.
I'm going to go through three of them and not in a great deal of detail.
But one of the ways of addressing this particular need having to do with
cord clutter is say, let's see, we can't fix all of the cord clutter, but
perhaps we can address some of cord clutter.
What we'll do is we'll develop a device that we can attach to the patient and
remove some of the cords that would be in the operating room
having to do with the patient and monitoring their vital signs.
As you see here,
one thing I want you to notice very carefully is with the exception of
this one little diagram in the corner here,
there are no numbers, there are no dimensions.
It's not even necessarily drawn to scale.
There's no indication of the material properties.
It is a conceptual diagram, a conceptual drawing.
If you look at the wheeled cart,
which is number two, you'll notice there's not even an indication of height, or
width, or weight, or diameter of the wheels.
The team is simply trying to identify conceptually,
what the device would look like.
It'd give you some idea of how the device would operate and
interact with its environment, and how clients or quote,
unquote, stakeholders and customers would actually interact with the device.
In terms of cord clutter, they came up with a third alternative, an integrated
operating table, in which everything is incorporated into the table.
On this particular one, they've shown
an anesthesia trolley is a separate item from the bed.
But for all intents and purposes, the operating table,
which may double as a bed or not, this is something you might want to consider,
would be completely independent of most everything else.
In other words, everything is contained within the table itself.
No dimensions, no weights, no voltages,
no current levels, it's a concept.
Now you can add more if you want to,
but we're trying to simply give enough information, so
someone that in their mind's eye can wrap their mind around, what's the concept?
How's this work?
How would it interact with the stakeholders?
Now, this team took another step and
said let's take a look at these stakeholder requirements.
And let's see if we can compare these requirements and test
them against the different devices that we have conceptually developed.
Now, note that the weighting here,
runs from 23 for is it effective as previous machines, or
previous devices, and this is what they started out with.
You'll see a difference in this in a few minutes, how this might change.
And when they looked at these three, and compared how these might
score against these weights, now remember we're using analytic hierarchy process,
which we talked about I think all the way back in week one.
When they scored all these, the Wheeled cart came up with a score of 80.99.
In other words, looking at these weights, looking at these device concepts,
this team believed that the Wheeled cart, based upon
these requirements, was the most attractive from their perspective.
Now, once we decided on the Wheeled cart and we have a conceptual drawing,
we can't really turn to an engineer and say, please now,
using your skills, go develop a detailed drawing.
With details of links, and widths, and weights,
and resistances, with no more than what we have,
so we need more information.
To do that, the team did two things.
They looked back at their client wants and needs.
They realized that some of those were very low-weighted, and to deal with 10 or
12 items was much more difficult to deal with as a smaller set.
And they chose to reduce the number of wants and needs and
to rescore, that's the reason you see the revised score here.
In other words, to reduce some complexity.
Now they could have left all of them in, but they chose to leave some out.
And I recommend that whenever you're using weighting criteria
[COUGH] that you
try to stay with something around seven.
You get much more than seven,
what happens is the scoring gets to be very close together in many cases.
And five is okay, but keep it to some small number.
And I like seven, it just seems to work.
Now, you look here, once you identified your wants and needs, then
we're looking at this from an engineering perspective, from a technical perspective.
And we're asking, what attributes, with whatever product we provide,
what attributes would it have to have in order to deliver or
to address these wants and needs?
So we're trying to link the wants and
needs to attributes of the physical product.
Now, we've shown here that the physical product,
one of the attributes has to do with accuracy.
One has to do with the speed of readings.
The number of cords reduced, the mobility.
Now, for some reason, people want to believe that this is almost automatic.
It's very, very simple to do, but it's not.
And one of the ways to determine what the attributes might be that you want to
use from an engineering perspective is to brainstorm them
with a group of people that are technically competent, and
technically capable of actually doing, or performing the design.
Now, when you do this what maybe long list of metrics,
before you start assigning units, just to save yourself some time, and energy and
effort, ask how does each one of these attributes relate to
each one of the requirements?
Each attribute must intersect with at least one requirement.
And if it doesn't, that's probably not an attribute
that's of any value to the product or the service that you're trying to develop.
There's no reason that one attribute can't be related to more than one requirement.
Take it a step further, there's no reason that one requirement can't have several
attributes related to it, but there must be at least a one to one row to column.
So when you do your brainstorming and you have a long list of attributes,
and you come back and you ask, how did the attributes, quote, unquote, map or
link to the requirements?
You may indeed determine that some of the attributes are not appropriate for
this particular design.
Or, there may be other attributes that
might be needed because we haven't related what the product or
service is going to do technically to what the customer said they wanted.
Now take a look over at the wants and needs, and
notice that these are expressed in what might be defined as customer voice.
They should be in a language that the customer or
the key stakeholders would use.
Now, once we have this linkage,
we still don't have enough information for the engineer to proceed.
So for each one of these attributes, we need to know the units of measure.
And the ones you see here, there's nothing wrong with these or right with these,
these are the units of measure this team, we think are appropriate.
Now, if you look over at Wireless, it just simply says yes or no.
Well, if Wireless had been a must, it wouldn't show here at all.
And what we might have is, instead of having wireless, we might have frequency,
or we might have range.
But for this point, the team is trying to decide, do we want wireless or not?
Or in another way of saying this,
if you take a look where it says Wireless, it's right at 6.36, it's not a must.
There is a chance that it might develop, a product or
service that doesn't have wireless.
Price is in US dollars, but we don't know what range.
In terms of size it says, appropriately sized.
That's going to have to be dealt with.
It's not nearly concise enough to be able to structure or
design a prototype of the card that we're talking about.
So, how do we determine what sizes we need?
What the dimension should be?
What the detail specification should be?
So, what we did here is for each one of these attributes,
and I'm only going to show you a portion of a table,
we're going to link the attributes now, to,
Some detailed specification.
Now, this one needs more work.
So we can think of it as being our first cut, because when you look at this,
we're going to talk about accuracy, and we said the unit is % error.
But what we'd really like to have in each one of these cells is the numerical value.
Let's go through this.
If you remember when we talked about the preliminary needs specification,
one of the attributes Well,
one of the requirements was to identify the competition for alternative methods.
For this particular one,
this team had identified eight current providers, I'm only showing you three.
So as we look at accuracy, and we look at the GE healthcare device or system,
it says that one test identified three out of three alarms correct,
leading competitors identified one out of three.
Okay, so we'd like to have high in terms of this particular
accuracy in terms of alarms.
LifeSync, one study found alarm accuracy to increase from 46% to 92% with
their lives.
Draeger had an unprecedented degree of accuracy.
And what this reflects when you see something like this is it
means that this particular team had not been able to find the level of detail or
level of specificity they really need.
Now, once you have identified five or six or
seven of these, maybe more, you ask, for accuracy,
what would be the ideal result for our product, or device, or service?
Ideally, we'd like it to be 100%, or improved over traditional methods.
And here, I would argue this needs to be a bit more specific than that.
The team, based upon their work, said,
if our device that we developed can't have a 95% confidence
in terms of accuracy, then we shouldn't do it at all.
So, in other words, what they've done is they've said, ideally and
what would be marginally acceptable for each one of these.
Now with the drawings, with the requirements, and
now with the specifications, the target specifications,
you should have enough information to be able to work with
designers that have the skill sets associated.
For example, if you have a device that's going to have electrical feedback used,
you certainly would like to have an electrical engineer,
perhaps an electrical computer engineer, working with you in your team.
So this will help you decide who you're going to need to build a prototype or
skill set and work with them to determine do you have enough information to move
ahead with the design and construction of your prototype.
The reason we need to build a prototype is at this point, we still don't know,
or we still don't have high confidence of what the specification should be in
terms of minimums and maximums and standard deviation, etc.
For each one of the attributes that our product or
service must address and address properly,
what we yield the kind of results that will allow
the stakeholders to achieve their requirements or their expectations.
The key question is, is there enough information from which to
prepare a detailed engineering drawing and go to prototype?
Now, to conclude, design is,
I think I have said in the past, is often misunderstood.
Design is not something magical.
Design is basically hard work.
What design does is it connects
the need and the solution with the product or service,
That meets the needs of the stakeholders and
can be developed and delivered by the providers or
by the innovative entrepreneurs that many of you are.
Again, thank you.
We look forward to seeing you next time.
Ознакомьтесь с нашим каталогом
Присоединяйтесь бесплатно и получайте персонализированные рекомендации, обновления и предложения.
Coursera делает лучшее в мире образование доступным каждому, предлагая онлайн-курсы от ведущих университетов и организаций.
© Coursera Inc., 2018 Все права защищены.
