High Tech Product Design and Rapid Prototyping ME221 MBA 290M INFOSYS 290'8

1
High Tech Product Design andRapid
PrototypingME221 - MBA 290M - INFOSYS 290.8

• Prof. Paul Wright, A. Martin Berlin Chair in
  Mechanical Engineering
• Chief Scientist of CITRIS _at_ UC Berkeley
• Co-Director of the Berkeley Wireless
  Research Center
• Co-Director of the Berkeley Manufacturing
  Institute
• Week 1a and 1b Introduction

2
Week 1a http//kingkong.me.berkeley.edu/Me221/in
dex.html

• Important to Review the Kingkong.me Site as
  early as possible It contains .
• Outline
• Syllabus
• Grading split
• Office Hours Monday Wednesday at 5pm in 2117
  Etcheverry
• Link to Our Labs shows prototyping equipment
  etc
• Resources for projects on
• Also see http//tinyos.net/

3
Class philosophy

• Learning By Doing
• The most important thing to realize today
  (especially if you are still in class-shopping
  mode)
• This is not a class with the answers in the back
  of the book
• There is intense group work --- like it or not,
  you will have to work closely with other people
• The class aims to give you the experience of
  developing new products in a big company like HP
  in a start-up

4
The main outcome

• From an emerging technology
• Develop idea(s) for a new consumer product
• Define its user scenario
• Do a design for the product
• Make a prototype casing
• Assemble the product
• Pitch it at a Trade Show on November 20th 2006
• (Attendance is mandatory at this Final Event)
• Develop a business plan for the product

5
This years chosen technology

• Telos motes that you will use as the inner
  brains to create a consumer product
• Previous years chosen technologies (FYI)
• RFID technology (04)
• Crossbow motes (03, 02, 01)
• Motorola TalkAbout Phones (00,99)
• Berkeley Wireless Center Products (98,97)
• WebCAM and BlimpCAMs (96,95)
• Head Mounted Displays (94)

6
The general experience in the past

• Understandable confusion at first
• As seen -- say at HP -- when the Next Big Thing
  is just emerging and people wonder what to do
  with it
• Nervousness in a group setting
• But by October 1st A nice idea!!
• Learning how to turn this nice idea into a
  manufactured product at reasonable cost by
  November 1st
• Enjoying the pitch on November 20th
• Enjoying developing the business plan and even
  getting excited enough to submit to Vertex or the
  Haas School competition

7
A pause for questions

• FAQs are on the Website
• But lets spend 5 10 mins now with a couple of
  questions and a discussion

8
More information on the motes as a product basis

• Etchnet as the installed network
• Your motes can connect with Etchnet
• Base-station mote
• Mobile mote

9
Hardware used with todays Telos 06 device
10
2 inch
Software (TinyOS) 25 motes along Floor 2 of
Etcheverry
11
Product ideas

• 1. Energy conservation
• 2. Health care
• General
• Specific illness
• Elder care
• 3. Rescue operations such as fire fighting
• 4. Environmental monitoring of water etc
• 5. Manufacturing and supply chains

12
Ideal examples will be related to energy and
medical applications
Monitoring
Control
13
Estrosense
Prototype
Tester Swab
ESTROSENSE
17.2
14
Future Healthcare in my own home ??
15
1. New Thermostat with touchpad shows price of
electricity in /kWhr expected monthly bill.
Automatic adjustment of HVAC price/comfort.
Appliance nodes glow-colors based on price.2.
New Meter conveys real-time usage, back to
service provider 3. Wireless beacons (smart
dust) throughout the house allow
for fine grained comfort/control
Vision Demand Response in a CA smart house
Incoming price signals

Appliance lights show price level appliances
powered-down
16
Product ideas

• 1. Energy conservation
• 2. Health care
• General
• Specific illness
• Elder care
• 3. Rescue operations such as fire fighting
• 4. Environmental monitoring of water etc
• 5. Manufacturing and supply chains

17
Week 1bMore product ideas, technology trends,
and costs

• InfogtgtBiogtgtNano challenges that will be able to
  use RFID tags

Paul Wright with thanks to Ravi Nemana
18
Product ideas

• 1. Energy conservation
• 2. Health care
• General
• Specific illness
• Elder care
• Hospital care
• With thanks to Ravi Nemana
• 3. Rescue operations such as fire fighting
• 4. Environmental monitoring of water etc
• 5. Manufacturing and supply chains

19
Problem How to build a good hospital?

• Background
• Architects design hospitals with 3D models, but
  they make assumptions about how humans interact
  with the confines of a facility.
• It is expensive and time consuming to follow
  humans around to see how they use a facility.
• When the facility is finally built, often up to
  10 of the cost is spent on tenant
  improvements.
• These are costs incurred (on a 300 M hospital,
  they can be as high as 30M) to change the
  facility to meet the needs of clinicians.
• Challenge
• Design a wearable, wireless device and sensor
  that can detect and log the interactions of the
  wearer. This data would then be used to model
  human activity in a new facility.

20
Problem How to spot infections in homebound
patients?

• Background
• If detected early, urinary tract infections
  (UTIs) are easy to fix
• If left untreated, they can lead to adverse
  events such as renal (kidney) failure and sepsis
  (a systemic infection that is much harder to
  treat).
• Homebound elderly frequently contract UTIs and
  fail to say
• The symptoms of UTIs include nonspecific things
  such as frequent urination and hardly noticeable
  pain.
• Research has shown that frequent urination (as
  compared to baseline) can be a good marker for a
  UTI.
• Challenge
• Design a wireless sensor that can log and track
  frequency of urination and report that
  information

21
(No Transcript)
22
(No Transcript)
23
Problem How to make ICUs safer?

• Background
• Over 100,000 lives are lost every year to adverse
  events in health
• Adverse events are unwanted incidents where harm
  resulted to someone receiving healthcare (which
  is supposed to be beneficial)
• ICUs are ripe areas because of the
  life-threatening nature of the care provided and
  the slim margins for error
• Errors often result in a greater degree of
  resources used to correct the error and help the
  patient. This increases costs.
• Many things contribute to an unsafe environment,
  including information and sensory overload,
  fatigue, poor data entry, illegible handwriting,
  poorly designed processes, and poor access to the
  patient
• Many patients are literally buried with cables,
  instruments, and devices so that clinicians have
  a hard time laying hands on the patient
• Challenge
• Make the ICU a safer environment by eliminating
  cables (see photos)

24
How to make ICUs safer?
Many sensors are tethered to a monitor and
displayed. Every device has a power source
battery in case of failure
25
Problem how to make hospitals safer?

• Background
• Some of the most difficult to treat infections
  originate in the hospital.
• These drug-resistant organisms prey on
  individuals with low ability to fight infections
  (the ill, immune-compromised, etc.)
• Clinical staff, instruments, and family members
  are often the vectors that spread infections
  within the hospital confines
• Usually, the patient, must be quarantined and
  treated, the room must be sanitized, and
  elaborate and expensive procedures must be put in
  place (scrubbing, gowns, mask, gloves for every
  contact with the patient)
• Hand washing is known to be an excellent
  deterrent to hospital acquired infections
• However, detection of the organism is often done
  after the spread.
• There is some evidence that detecting an organism
  in the water used to wash hands may alert medical
  personnel to the presence of drug resistant
  organisms.
• Challenge
• Design a wireless sensor and system to detect
  drug resistant organisms in the waste water of
  hospital sinks.

26
Problem Instrumentation for remote patient
management

• Background
• Remote patient management (RPM) is a field that
  allows physicians to care for patients from a
  distance.
• RPM assumes that the processes of diagnosis and
  treatment can be decoupled and that one need not
  be within proximity of a patient.
• To do this, information and communications
  technology must extend the senses of the
  clinician to allow for diagnosis to be made from
  a distance.
• Clinicians use sight, sound, touch and smell to
  make a diagnosis.
• Otitis media is an easily treated common ear
  infection in infants and children. To diagnose
  this, one needs to see a red eardrum and display
  symptoms of fever, irritability, scratching at
  the ears, etc.
• To manage OM from a distance, one needs the image
  of the eardrum and the symptom interview to come
  to a diagnosis.
• Challenge
• Create a wireless sensor that can reliably take a
  picture of the eardrum and transmit it to a cell
  phone or other platform.
• Keep in mind that the inside of the ear is
  normally dark.

27
Macro to Micro Computers (needed for such a
design inside the ear)
Stand alone computers 1960s

Connected computers 1980s
Distributed computers 2000s
Smart Dust
Vast reduction in cost, but additional capability
1940s
2000s
Year
Adapted from Various Sources E.g. G. Bell, R.
Newton, J, Rabaey, D. Culler, DR research
28
1983- Hewlett-Packard scientist Joel Birnbaum
uses the expression pervasive computing in a
research paper.
1988- Xerox researchers begin building tabs,
pads and boards with embedded networking.
1991- Mark Weiser publishes The Computer for the
21st Century , envisions a world of ubiquitous
but invisible computers.
1993- Internet goes mainstream with first
graphical Web browser.
1996- The Pilot (PalmPilot) makes computing easy,
cheap and popular.
1998- Digital mobile phones make wide area
connectivity portable and cheap.
2000 Local area wireless networking
technologies are standardized
gtgt2010 Cheap Sensors, omnipresent wireless and
small handhelds create a new platform for
pervasive computing (e.g. in hospitals .
29
Question

• Describe a new products that you have seen that
  approach this embodiment idea
• E.g. Home kits for..
• Blood pressure monitoring (but not automatic)
• Diabetes monitoring kits (also not automatic)
• Pls send URL to me pwright_at_me.berkeley.edu

30
The basic cost of a product

• Begin with outsourced cost to manufacture (Bill
  of materials Assembly costs)
• Cost to manufacture (BOM Assembly) X
• Other costs (Sales and Marketing 0.5x (but
  colgate palmolive 4x) Overhead and/or Non
  Recoverable Eng., Gen admin. 0.5x
• Therefore X 0.5X 0.5X y 0.2y (profit
  whereas at Intel it might be 1.0) Z where Z2.2
  to 2.5X
• Z contractor or OEM (very high volume, no sales
  channel)
• 1.5x distributor (Greybar)
• 1.5x outlet (Home Depot)
• Thus price to the person on the street is 3or4X

31
We will go back to this again

• Will your company make motes?
• Hard
• Will your company use motes for a product that
  assumes an infrastructure?
• Challenging to work with big government (or even
  BART)
• Will you be a design team like IDEO or Frog
  design that creates a product around a
  base-station and mobile system that can be used
  in a defined setting?...
• More likely to work

32
Personal homework this weekend

• Realize this class is learning-by-doing with
  group work
• Make sure it suits you
• Its not an undergraduate cramming class
• No answers in the back of the book
• No good reply to what is on the next test
• No good reply to what does the professor want
  from me
• Except be creative, come up with a great product
  that will impress the market as well as the
  judges on November 20th 2006
• Its more like real life than school

33
Group homework this weekend

• Over Labor Weekend, chat with other friends, look
  around you duringbiking, shopping, have fun
  thinking about something that has always bugged
  you Youre always losing something, need
  better information,
• How might motes map out new product solution
• Identify specific area
• Identify cost barriers
• Think about -- return on investment (ROI) your
  friends will have plenty of ideas..But will it
  make you a millionaire?
• Think about -- GO or NO GO
• Next week after Labor Weekend discuss with us
  all in class
• Week after that work with team

34
The challenge of the price point to the consumer

• 0.99
• 9.99
• 19.99
• 49.99
• 99.99
• 249.99
• 499.99
• All these Price Points mean Cost Points 3 to
  4 times less

35
Summary Product ideas

• 1. Energy conservation
• 2. Health care
• General
• Specific illness
• Elder care
• 3. Rescue operations such as fire fighting
• 4. Environmental monitoring of water etc
• 5. Manufacturing and supply chains

36
Backup Slides
37
Manufacturing issues Number of Levis Jeans
factories in the US In 1980 60 In 2004 ??
How many
38
Number of Levis Jeans factories in the US In
1980 60 In 2006 Zero More Questions !! In a
24 year span would it be the same in
China!!? Where else will US and Euro countries
go for labor?
39
From The Economist March 6th
Sewing Machine Operators Minimum Salaries US
minimum 893.20 Honduras 139.00
Guangdong 63.75 Bangladesh only 18.53
per month !!
40
Evidence Based Medicine
41
Obvious Cost Barriers to Overcome

• Cost of mote technology itself
• Cost of applying the motes to
• people,
• devices,
• equipment,
• merchandise,
• Etc..

42
Less obvious costs that helpReturn on Investment
(ROI)

• Cost of losing a life avoided (shadow cost)
• Legal penalty for wrong treatment
• Lower costs in inventory management of drugs
• Lower costs of equipment management or
  maintenance
• Unique IDs can reduce hospital administrative
  costs and if implemented by local government, can
  tie into other social services