Note on Engineering Product Design
M I T S L O A N C O U R S E W A R E > P. 1
Note on “Engineering” Product Design
John R. Hauser
In the “Note on Product Development” we explored the use of the Lens’
model in product design. This note presents an example prescriptive method to
“engineer” product design based on the Lens’ model. In particular, we recognize
that customers choose products based on how well the products fulfill their
perceived needs. Product development teams must select product features to
fulfill those perceived needs.
Why is this so hard? Consider a liquid dishwashing detergent (for
washing dishes by hand). It is basically a chemical product. Customers buy it to
clean their dishes. But what do clean dishes mean and how does the customer
judge that their dishes are clean? It is unlikely that they use a magnifying glass or
an instrument to measure the light reflected from the dishes, although an engineer
might use those instruments to test a dishwashing product’s performance. It is
also unlikely that the customer will read and understand the chemical ingredients.
More likely, the customer will use some subjective means to determine “clean.”
This might mean holding them up to the light or it might simply mean that there is
M I T S L O A N C O U R S E W A R E > P. 2
no noticeable dirt on the dishes – a minimum requirement. Alternatively, the
customer might derive peripheral cues, such as the clarity of the water in the sink
(after washing dishes) or the amount and type of bubbles that are still around
when the dishes are done and ready to be rinsed. The scent of the dishwater or of
the dishes might be another cue to cleanliness.
But “clean” might not be the only perceived need. The customer might
care about the ease of use, the “feel” of the water while washing the dishes, the
scent of the liquid, the ease of storing the bottle (or package), whether or not the
washing causes the dishes to deteriorate, or whether the liquid imparts a perceived
taste to the dishes. It is not enough to engineer the best-cleaning liquid. The
product development team must engineer the entire customer experience. This
includes the liquid itself as well as the packaging and any advertising image.
Consider another example – engineering a customer-service center to
provide great service. The development team may design the physical space,
select the telephonic equipment, and design protocols. The development team
may also design a personnel policy, training, and a monitoring/reward system.
Consider the monitoring/reward system. One major financial service provider
determined that customers did not like to wait. They instituted metrics and
rewarded the service providers to minimize the number of rings before answering,
avoid transferring customers, and answer the customers’ questions as rapidly as
possible. Unfortunately, the service providers figured out the system – answer the
phone quickly, avoid a transfer even if the service provider did not know the
answer, and get off the phone as quickly as possible. In fact, the metric was
improved by giving incomplete and unhelpful answers.
The financial services firm responded. Service providers were now given
incentives to stay on the phone until the customer got an answer. One metric was
the number of minutes per hour that the service provider was on the phone. The
service providers responded. They were never off the phone. Breaks were taken
while the customer was on hold.
Ultimately, the financial services firm talked to the service providers and
understood their needs. Service providers were not happy with gaming the
M I T S L O A N C O U R S E W A R E > P. 3
system. They wanted to serve customers but felt that the metrics prevented them
from doing so. In response, the firm began using more qualitative metrics based
on monitoring and on customer satisfaction scores. Service providers started to
maximize the customers’ perceived needs – the customers wanted their questions
answered correctly. Efficiently was a secondary criterion.
Many other example abound. Fulfilling customer needs is important to the
success of a new product, but it is difficult to achieve. It is especially difficult in
a complex product such as an automobile or a high-end copier. An automobile
may take 1,000 person-years to design – millions of decisions need to be made.
Even an office copier may require close to 10,000 critical engineering decisions.
We want every one of those decisions to be focused on the customer.
The House of Quality
There are many means to match product (or service) features to perceived
customer needs. For small numbers of features we often use “conjoint analysis,”
a method we cover later in the course. In this note we provide one example
method that has been used for many types of products varying from laundry
detergents (P&G) to services (Pacific, Gas, and Electric, Inc.) to automobiles
(Toyota, General Motors, Ford). It has even been used for movie theaters and off-
shore drilling equipment.
Figure 1 illustrates a stylized House of Quality (HOQ) completed by
Puritan-Bennett when they were redesigning a medical instrument called a
spirometer (measures lung capacity). We cover the details of this case in class.
The HOQ begins with a formal study of perceived customer needs called the
voice of the customer.1 The voice of the customer identifies customer needs such
as “the product is affordable” or “the product is easy for the physician to hold
while taking measurements on a patient.” In the Puritan-Bennett case, interviews
with physicians, technicians, nurses, and patients identified 25 strategic customer
needs. These are listed on the left side of the house.
See “Note of the Voice of the Customer.”
M I T S L O A N C O U R S E W A R E > P. 4
But all needs are not equally important. The customers would much
prefer that some needs be fulfilled even if it means that other needs are not
fulfilled as well. The voice of the customer measures the importance of each
need. In Figure 1 these importances are listed next to the customer needs. In
addition, the HOQ lists on the right how each existing product (PB, SM, or WA –
three manufacturers) fulfill the customer needs. For example, WA does
extremely well on the important customer need of affordability, while PB does
poorly. If the design team can fulfill this need at a reasonable cost it is of high
(25 strategic needs) Relationships
Product is affordable 100 between
Easy to hold 72
Easy to clean 86 PB SM WA
Convenient-sized output 86 Customer
Costs and Feasibility
Figure 1. Illustrative House of Quality
The remainder of the HOQ is self-explanatory. The product development
team lists design features at the top of the house. For spirometers a design feature
might be the size, shape, or color of the instrument. It might be the means by
which data is stored or the means by which data is displayed to the physician. It
might include software, hardware, or even packaging features.
M I T S L O A N C O U R S E W A R E > P. 5
The team them considers each high priority customer need, such as
“product is affordable,” and, in the center of the house, indicates how each design
feature affects the fulfillment of that need. For example, special plastics may
make the spirometer easy to clean (positive relationship) but expensive to produce
(negative relationship with “product is affordable”). Some teams indicate the
relationships with symbols ( , , ), others with numbers (±9, ±3, ±1), and still
others with simply + or –. The HOQ is a guide; any quantification must be
tempered with qualitative judgment.
The HOQ is then completed with costs and feasibility and, possibly,
metrics (engineering measures) at the bottom of the house.
Early applications used large numbers of customer needs and design
features, but the HOQ became tedious. The HOQ evolved to a representation that
was more “just in time.” When the team considers a customer need, it fills in the
design features to which that need is linked.
Communication Among Product Development Team Members
The HOQ is a management tool designed to enhance communication. It
does not automate product design. The team must still consider customer needs
and use its creativity and its firm’s strategic resources to develop a product that is
profitable (and will remain so under competitive threat).
In the early days of the HOQ, MIT Sloan researchers studied its
implementation at Ford Motor Company. Two teams were chosen. Each was
working on a similar, but different, component of a new automobile and each
team was otherwise similar in skills and team members. Both teams reported to
the same managers. One team used the HOQ; another team used Ford’s standard
phase review process. Over the course of the product development cycle, the
MIT Sloan researchers measured the amount and type of communication among
team members. The results are given in Figure 2.
M I T S L O A N C O U R S E W A R E > P. 6
0 2 4 6 8 10
Phase-Review House of Quality
* Significant at the 0.05 level.
Communications per week
Figure 2. Communication at Ford
The light grey bars indicate the communications per week measured for
the team using the HOQ. The dark red bards indicate the communications per
week measured by the team using the phase-review process. Overall, there was
significantly more communication by members of the HOQ team. Deeper
analysis revealed that the HOQ focused its communications within the team –
both within functions and between functions on these interfunctional teams. The
only type of communication reduced by the HOQ was to management. Even
deeper analysis showed that the phase-review team engaged in significantly more
“up-over-down” communication. For example, an engineer at Ford might
communicate a design change to his/her manager who would communication that
design change to a manager at a supplier who would, in turn, communicate the
design change to an engineer at the supplier. In the HOQ team the Ford engineer
communicated directly to the engineer at the supplier.
Figure 2 is just one example on the success of the various means to
enhance communication among product development team members on the
relationship between perceived customer needs and engineering design features.
Figure 2 is based on the HOQ, but there are many other ways to effect
communication. For example, Wind, et. al. (1989) provide an excellent example
M I T S L O A N C O U R S E W A R E > P. 7
of how conjoint analysis was used to design “Courtyard by Marriott” using
conjoint analysis to link the features of the hotel to perceived customer needs.
Green, Krieger, and Wind (2004) provide another example where conjoint
analysis was used to design the EZPass (FastLane) system. (See “Note on
Conjoint Analysis” for more detail on how to collect data from consumers and
how to use conjoint analysis in marketing management.)
The important lesson for 15.810 is that such communication must occur if
successful products are to be designed.
Akao, Yoji (1987), Quality Deployment: A Series of Articles, (Lawrence, MA:
G.O.A.L., Inc.) Glen Mazur, trans.
Griffin, Abbie, and John R. Hauser (1992a), "Patterns of Communication Among
Marketing, Engineering and Manufacturing -- A Comparison Between Two
New Product Teams," Management Science, 38, 3, (March), 360-373.
Green, Paul E., Abba Krieger, and Jerry Wind (2004), “Thirty Years of Conjoint
Analysis: Reflections and Prospects,” in Jerry Wind and Paul E. Green,
eds., Market Research and Modeling: Progress and Prospects, a Tribute to
Paul E. Green, (New York, NY: Kluwer Academic Publishers).
Hauser, John R. and Don P. Clausing (1988), "The House of Quality," Harvard
Business Review, 66, 3, (May-June), 63-73.
_____ and Vithala Rao (2004), “Conjoint Analysis, Related Modeling, and
Applications,” Marketing Research and Modeling: Progress and Prospects,
Jerry Wind and Paul Green, Eds., (Boston, MA: Kluwer Academic
Krieger, Abba, Paul E. Green, Leonard Lodish, J. D’Arcangelo, C. Rothey, and P.
Thirty (2003), "Consumer Evaluation of 'Really New' Services: The
Traffic Pulse System," Journal of Marketing Services, 17, April.
Wind, Yoram, Paul E. Green, Douglas Shifflet, and Marsha Scarbrough (1989),
"Courtyard by Marriott: Designing a Hotel facility with Consumer-Based
Marketing Models," Interfaces, 19, (January-February), 25-47.
- Related pdf books
- Effect of spherulites on the mechanical properties of nylon 66
- Low-Cost Prosthetic Hand Fabrication Process
- Blackness and Blood: Interpreting African American Identity
- Opportunity Mars Rover mission: Overview and selected results ...
- Kagan, Shelly. The Geometry of Desert 1 - MIT - Massachusetts ...
- Magnetic Endotracheal Tube Imaging Device
- Computing Local Surface Orientation and Shape from Texture for
- 13.021 Marine Hydrodynamics, Fall 2004 - MIT - Massachusetts ...
- Precursor mechanism for interaction of bulk interstitial ...
- Mammalian mutagenesis using a highly mobile somatic Sleeping ...
- Techniques for minimizing error propagation in decision feedback detectors for recording channels -
- An Exchange of Ideas in Pabal: Defining Development - Spring ...
- On the Complexity of Two-Player Win-Lose Games
- A Sample Game (Part 1 of 4) - MIT - Massachusetts Institute ...
- Adrienne Rothschilds CV | 1 Adrienne Rothschilds EDUCATION UW ...
- Behavioral/Systems/Cognitive Synchronous, Focally Modulated ...
- The Rational Basis of Representativeness