SlideShare a Scribd company logo
Application of Agile and Lean
Methods to Hardware Product
Development
By: Vincent Spena
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 1
Kanban Approach
1. In a “factory”, the production steps are
repeatable and each pre-defined step adds value
2. Kanban is used to “measure for the purpose of
control”
• It can be implemented in many ways including
the use of tokens or computerized forms
3. The sample Kanban table to the left provides
• information about how items are flowing
though the process steps
• information about the throughput of each
station (when time is factored in)
Great Lakes Regional Conference 2012
Copyright © 2012 by : Vincent Spena
Published and used by INCOSE with permission 2
Station A
Station C
Station D
Station B
Station G
Station F
Station E
Line “value”
Items
A B C D E F G
1 50 45 40 40 35 20 20 15
2 75 75 70 65 40 15 10 10
3 60 45 45 45 30 25 20 15
4 45 45 45 40 35 25 15 15
5 80 80 40 40 35 35 25 15
Sample Observations:
• Line 1 has 5 items that have not been put into the
process
• Line 2 seems have some throughput issues at station E
because 25 items are in queue
• Line 3 might be “overstaffed” for station B and station C.
Maybe staffing can be redeployed to station A
What is happening?
1. Each item must be processed with a
predetermined number of steps --->
A+B+C+D+E+F+G (Not necessarily in sequence)
2. Each step adds a “value” to the item. The
“amount” of value for each step may not be
known, but the measurement provides
information that highlights if the value has
been added or if it still needs to be added.
3. A sequence may not be necessary, but if it’s set
up to do so, the value in “Station D” cannot be
added without the unit having A+B+C
Great Lakes Regional Conference 2012
Copyright © 2012 by : Vincent Spena
Published and used by INCOSE with permission 3
Station A
Station C
Station D
Station B
Station G
Station F
Station E
Line “value”
Items
A B C D E F G
1 50 45 40 40 35 20 20 15
2 75 75 70 65 40 15 10 10
3 60 45 45 45 30 25 20 15
4 45 45 45 40 35 25 15 15
5 80 80 40 40 35 35 25 15
Assumptions:
• The operation in each station may be short and
repeatable – (ie visible to the operator)
• Each step is assumed to be perfect -- the mechanism to
detect and prevent faults in lean systems is left to jidoka,
and the operator is an integral part of keeping faults out
of the system (Ie “Built in Quality”)
Iterative and Incremental
Development - IID
IID Is often associated with Software Development, but consider what
happens in the construction of a new building :
1. The plans of the TO-BE building is defined
2. The skeletal framework is constructed
3. The surrounding structure is built up on the skeletal framework
4. The details and final touches are placed in the structure
In all cases the following items are important:
 A plan with high level requirements of the final outcome
 Defined steps and defined requirements for each step. ( Ie: Building frame
with 15 floors, concrete structure for 15 floors, 30 rooms in each floor, etc)
 Each step requires the previous step to be sufficiently complete before the
next step can begin
 Refinements can be made along the way as long as they conform to the
original plan. (Ie 29 rooms on the 15th floor with one room being double the
size of a normal room).
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 4
1
2
3
4
Anatomy of complex product development
• Product development can be thought of many separate development efforts
• Each team develops different parts of the product at different rates depending on difficulty,
resources, knowledge, and dependencies such as part availability, build schedules, and knowledge
sharing
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 5
Control Panel
GPS Module
Bluetooth Module
Chassis
Software
Certifications
Core
Features
Amplifier
Receiver
Oscillator
Processor
Peripherals
These teams share
a common board
and integrated
build plan
These teams can
work somewhat
independently
Time
Applied Kanban and IID
Great Lakes Regional Conference 2012
Copyright © 2012 by : Vincent Spena
Published and used by INCOSE with permission 6
Activity A
Activity C
Activity D
Activity B
Activity G
Activity F
Activity E
P1
P2
P3
1. A Prototype is a grouping of all the activities
that have been performed.
2. Like in IID, prototypes must have planned
design targets (requirements to be met).
Design targets provide a frame of reference
for the whole project
3. IID provides an alignment point for many
teams
• Without IID targets, teams try to
dictate project changes based on their
needs instead of aligning to project
goals.
4. Requirements are of two varieties:
• Requirements that are internal to
development (Not counted because it
cannot be defined- see Assertion #1)
• Requirements that are to be verified
for the project (IID targets)
Assertion #1: In development, a station is a type of
activity like testing a circuit or collecting part
specifications. Therefore the sequence and quantity of
development activities is not predictable.
Assertion #2: Since the quantity of activities for a
prototype cannot be known: A comparable means to
gauge the effectiveness of activities is to measure the
requirements that have been met for that prototype
Application of IID and Kanban
Prototype quantity (plan) and Requirements to be met by each team at each prototype must be defined during the
project definition stage.
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 7
Control Panel
GPS Module
Bluetooth Module
Chassis
Software
Certifications
Core
Features
Amplifier
Receiver
Oscillator
Processor
Peripherals
These teams share
a common board
and integrated
build plan
These teams can
work somewhat
independently
P1 P2 P3Project
Definition
First Layer
Requirements
Complete
Second Layer
Requirements
Complete
Third Layer
Requirements
Complete
Start
Applied Kanban and IID
Great Lakes Regional Conference 2012
Copyright © 2012 by : Vincent Spena
Published and used by INCOSE with permission 8
Activity A
Activity C
Activity D
Activity B
Activity G
Activity F
Activity E
P1
P2
P3
The # of Prototype is predictable
The # of requirements to me met is predictable
The requirements met in the Prototype is a direct measure
of “Value” added to the design.
Engineer or
Team
Project
Requirements
P1
Pass
P1
Fail
P2
Pass
P2
Fail
P3
Pass
P3
Fail
Transmitter 150 45 8 107 12 150 1
Receiver 120 60 5 90 9 120 0
GPS 300 35 20 205 6 300 0
Control Panel 105 45 0 45 3 105 0
Sample Observations:
• It does not make sense to test for P2 and P3 functionality if
the product is at P1 maturity (should be expected but “test
everything” is not always seen as waste)
• The GPS subsystem may be having some issues because
>36% of “expected to pass” requirements were not met
• The Control Panel subsystem seems to have had a delay in
development/testing relative to the other subsystems.
Additionally: Using similar methodologies, the number of fault
items, pass items, time between protos, and other information
can be captured to gauge proto effectiveness against targets.
Note: Jidoka is much more difficult in development because steps
are difficult to visualize as being related. Reviews and other fault
avoidance mechanisms like “best practices” are still needed.
Sample Charts –
Overall Prototype Effectiveness
• A Prototype effectiveness chart clearly
shows performance relative to plan
• The % Fail line can be used to align to
trend information between similar
projects
• The chart is similar to a “burn up”
chart and the same data can be used
to create such a chart (shown below)
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 9
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
0
50
100
150
200
250
P1 P2 P3 P4
Prototype Effectiveness
FAIL Not tested PASS SW % FAIL
0%
20%
40%
60%
80%
100%
120%
P1 P2 P3 P4
EarnedValue
Ramp Up
Actual Planned
Sample Charts –
Prototype Effectiveness by team
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 10
0
10
20
30
40
50
60
70
80
90
100
ID ME RX TX BT GPS ID ME RX TX USB BT GPS ID ME RX TX USB BT GPS ID ME RX TX USB
P1 P2 P3 P4
Fail Pass SW Not Tested
Summary and Conclusions
• Kanban and IID can be applied to Product Development and
allows measurement for control of development activities
• Implementation of Kanban and IID in Product Development
requires:
– A plan that defines number of steps (Protos) and the
requirements to be met at each step
– A measurement of actual vs planned
– Mechanisms to prevent faults from propagating because
jidoka is not as straightforward during development
• Metrics based on Kanban and IID can allow near-real time
information about the progress of development
Great Lakes Regional Conference 2012
Copyright © 2012 by: Vincent Spena
Published and used by INCOSE with permission 11

More Related Content

PPTX
Product Design Using Solidworks
PDF
Path to Production: Value Stream Mapping in a DevOps World
PPTX
Secrets of Value Stream Mapping for Future State
DOC
K Chandrasekaran
PPTX
Zeb Pitch
DOCX
1 JimRinehart ResumeQE 15 12-4-15
PDF
RACI Team Practice
PDF
Continuous testing & devops with @petemar5hall
Product Design Using Solidworks
Path to Production: Value Stream Mapping in a DevOps World
Secrets of Value Stream Mapping for Future State
K Chandrasekaran
Zeb Pitch
1 JimRinehart ResumeQE 15 12-4-15
RACI Team Practice
Continuous testing & devops with @petemar5hall

What's hot (20)

PPTX
Advanced Work Packaging - CCT Solutions at a Glance
DOC
Civitano_resume_2017
PPT
Engineering Services Capabilities
DOC
ResumeSagli_Nov14
PPTX
Exploratory testing and Dev Ops - best friends?
PDF
Agile Introduction
PDF
Accounting for non functional and project requirements - cosmic and ifpug dev...
DOC
CASQUEJO, Arnold_Aug 2015
PDF
AutoSpice Agile Hand in Hand
PPTX
Advanced Work Packaging - Real Life Implementation
PPTX
ALM 101: An introduction to application lifecycle management
PPTX
Advanced Work Packaging (AWP) - Reduction to Practice
DOC
Resume_LI
PDF
Agile ALM Tool Comparison
PPTX
Agile conference 2013
PDF
OpenERP - Project Methodology
PDF
Jason E Stephens’ 2020 Resume/CV
PDF
Performance Testing in the Agile Lifecycle
PDF
John G Gizowski PMP
DOC
abhi resume
Advanced Work Packaging - CCT Solutions at a Glance
Civitano_resume_2017
Engineering Services Capabilities
ResumeSagli_Nov14
Exploratory testing and Dev Ops - best friends?
Agile Introduction
Accounting for non functional and project requirements - cosmic and ifpug dev...
CASQUEJO, Arnold_Aug 2015
AutoSpice Agile Hand in Hand
Advanced Work Packaging - Real Life Implementation
ALM 101: An introduction to application lifecycle management
Advanced Work Packaging (AWP) - Reduction to Practice
Resume_LI
Agile ALM Tool Comparison
Agile conference 2013
OpenERP - Project Methodology
Jason E Stephens’ 2020 Resume/CV
Performance Testing in the Agile Lifecycle
John G Gizowski PMP
abhi resume
Ad

Similar to Vincent Spena - Agile and Lean Methods for Hardware Product Development (20)

PPT
Kanban, Flow and Cadence
PPTX
Agile product development and project management with Kanban
PPT
Patton kanban
PPTX
Scrumban Demystified
PPT
Patton kanban 1
PPT
kanban.ppt
PPTX
Introduction to Agile Software Development
PDF
Kanban
PDF
Agile & Lean & Kanban in the Real World - A Case Study
PDF
Just-in-time development with kanban - JAX London 2011
PDF
Agile & kanban in Coordination
ODP
Intro to Agile and Lean Software Development
PDF
Driving Innovation with Kanban at Jaguar Land Rover
PDF
Kanban and Government
PPT
Lean Design - Literature survey
PPT
Agile overview
PDF
Kanban
ODP
Agile methods training
PPTX
Patton kanban fr
PPTX
Agile in Medical Software Development
Kanban, Flow and Cadence
Agile product development and project management with Kanban
Patton kanban
Scrumban Demystified
Patton kanban 1
kanban.ppt
Introduction to Agile Software Development
Kanban
Agile & Lean & Kanban in the Real World - A Case Study
Just-in-time development with kanban - JAX London 2011
Agile & kanban in Coordination
Intro to Agile and Lean Software Development
Driving Innovation with Kanban at Jaguar Land Rover
Kanban and Government
Lean Design - Literature survey
Agile overview
Kanban
Agile methods training
Patton kanban fr
Agile in Medical Software Development
Ad

Vincent Spena - Agile and Lean Methods for Hardware Product Development

  • 1. Application of Agile and Lean Methods to Hardware Product Development By: Vincent Spena Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 1
  • 2. Kanban Approach 1. In a “factory”, the production steps are repeatable and each pre-defined step adds value 2. Kanban is used to “measure for the purpose of control” • It can be implemented in many ways including the use of tokens or computerized forms 3. The sample Kanban table to the left provides • information about how items are flowing though the process steps • information about the throughput of each station (when time is factored in) Great Lakes Regional Conference 2012 Copyright © 2012 by : Vincent Spena Published and used by INCOSE with permission 2 Station A Station C Station D Station B Station G Station F Station E Line “value” Items A B C D E F G 1 50 45 40 40 35 20 20 15 2 75 75 70 65 40 15 10 10 3 60 45 45 45 30 25 20 15 4 45 45 45 40 35 25 15 15 5 80 80 40 40 35 35 25 15 Sample Observations: • Line 1 has 5 items that have not been put into the process • Line 2 seems have some throughput issues at station E because 25 items are in queue • Line 3 might be “overstaffed” for station B and station C. Maybe staffing can be redeployed to station A
  • 3. What is happening? 1. Each item must be processed with a predetermined number of steps ---> A+B+C+D+E+F+G (Not necessarily in sequence) 2. Each step adds a “value” to the item. The “amount” of value for each step may not be known, but the measurement provides information that highlights if the value has been added or if it still needs to be added. 3. A sequence may not be necessary, but if it’s set up to do so, the value in “Station D” cannot be added without the unit having A+B+C Great Lakes Regional Conference 2012 Copyright © 2012 by : Vincent Spena Published and used by INCOSE with permission 3 Station A Station C Station D Station B Station G Station F Station E Line “value” Items A B C D E F G 1 50 45 40 40 35 20 20 15 2 75 75 70 65 40 15 10 10 3 60 45 45 45 30 25 20 15 4 45 45 45 40 35 25 15 15 5 80 80 40 40 35 35 25 15 Assumptions: • The operation in each station may be short and repeatable – (ie visible to the operator) • Each step is assumed to be perfect -- the mechanism to detect and prevent faults in lean systems is left to jidoka, and the operator is an integral part of keeping faults out of the system (Ie “Built in Quality”)
  • 4. Iterative and Incremental Development - IID IID Is often associated with Software Development, but consider what happens in the construction of a new building : 1. The plans of the TO-BE building is defined 2. The skeletal framework is constructed 3. The surrounding structure is built up on the skeletal framework 4. The details and final touches are placed in the structure In all cases the following items are important:  A plan with high level requirements of the final outcome  Defined steps and defined requirements for each step. ( Ie: Building frame with 15 floors, concrete structure for 15 floors, 30 rooms in each floor, etc)  Each step requires the previous step to be sufficiently complete before the next step can begin  Refinements can be made along the way as long as they conform to the original plan. (Ie 29 rooms on the 15th floor with one room being double the size of a normal room). Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 4 1 2 3 4
  • 5. Anatomy of complex product development • Product development can be thought of many separate development efforts • Each team develops different parts of the product at different rates depending on difficulty, resources, knowledge, and dependencies such as part availability, build schedules, and knowledge sharing Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 5 Control Panel GPS Module Bluetooth Module Chassis Software Certifications Core Features Amplifier Receiver Oscillator Processor Peripherals These teams share a common board and integrated build plan These teams can work somewhat independently Time
  • 6. Applied Kanban and IID Great Lakes Regional Conference 2012 Copyright © 2012 by : Vincent Spena Published and used by INCOSE with permission 6 Activity A Activity C Activity D Activity B Activity G Activity F Activity E P1 P2 P3 1. A Prototype is a grouping of all the activities that have been performed. 2. Like in IID, prototypes must have planned design targets (requirements to be met). Design targets provide a frame of reference for the whole project 3. IID provides an alignment point for many teams • Without IID targets, teams try to dictate project changes based on their needs instead of aligning to project goals. 4. Requirements are of two varieties: • Requirements that are internal to development (Not counted because it cannot be defined- see Assertion #1) • Requirements that are to be verified for the project (IID targets) Assertion #1: In development, a station is a type of activity like testing a circuit or collecting part specifications. Therefore the sequence and quantity of development activities is not predictable. Assertion #2: Since the quantity of activities for a prototype cannot be known: A comparable means to gauge the effectiveness of activities is to measure the requirements that have been met for that prototype
  • 7. Application of IID and Kanban Prototype quantity (plan) and Requirements to be met by each team at each prototype must be defined during the project definition stage. Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 7 Control Panel GPS Module Bluetooth Module Chassis Software Certifications Core Features Amplifier Receiver Oscillator Processor Peripherals These teams share a common board and integrated build plan These teams can work somewhat independently P1 P2 P3Project Definition First Layer Requirements Complete Second Layer Requirements Complete Third Layer Requirements Complete Start
  • 8. Applied Kanban and IID Great Lakes Regional Conference 2012 Copyright © 2012 by : Vincent Spena Published and used by INCOSE with permission 8 Activity A Activity C Activity D Activity B Activity G Activity F Activity E P1 P2 P3 The # of Prototype is predictable The # of requirements to me met is predictable The requirements met in the Prototype is a direct measure of “Value” added to the design. Engineer or Team Project Requirements P1 Pass P1 Fail P2 Pass P2 Fail P3 Pass P3 Fail Transmitter 150 45 8 107 12 150 1 Receiver 120 60 5 90 9 120 0 GPS 300 35 20 205 6 300 0 Control Panel 105 45 0 45 3 105 0 Sample Observations: • It does not make sense to test for P2 and P3 functionality if the product is at P1 maturity (should be expected but “test everything” is not always seen as waste) • The GPS subsystem may be having some issues because >36% of “expected to pass” requirements were not met • The Control Panel subsystem seems to have had a delay in development/testing relative to the other subsystems. Additionally: Using similar methodologies, the number of fault items, pass items, time between protos, and other information can be captured to gauge proto effectiveness against targets. Note: Jidoka is much more difficult in development because steps are difficult to visualize as being related. Reviews and other fault avoidance mechanisms like “best practices” are still needed.
  • 9. Sample Charts – Overall Prototype Effectiveness • A Prototype effectiveness chart clearly shows performance relative to plan • The % Fail line can be used to align to trend information between similar projects • The chart is similar to a “burn up” chart and the same data can be used to create such a chart (shown below) Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 9 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 0 50 100 150 200 250 P1 P2 P3 P4 Prototype Effectiveness FAIL Not tested PASS SW % FAIL 0% 20% 40% 60% 80% 100% 120% P1 P2 P3 P4 EarnedValue Ramp Up Actual Planned
  • 10. Sample Charts – Prototype Effectiveness by team Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 10 0 10 20 30 40 50 60 70 80 90 100 ID ME RX TX BT GPS ID ME RX TX USB BT GPS ID ME RX TX USB BT GPS ID ME RX TX USB P1 P2 P3 P4 Fail Pass SW Not Tested
  • 11. Summary and Conclusions • Kanban and IID can be applied to Product Development and allows measurement for control of development activities • Implementation of Kanban and IID in Product Development requires: – A plan that defines number of steps (Protos) and the requirements to be met at each step – A measurement of actual vs planned – Mechanisms to prevent faults from propagating because jidoka is not as straightforward during development • Metrics based on Kanban and IID can allow near-real time information about the progress of development Great Lakes Regional Conference 2012 Copyright © 2012 by: Vincent Spena Published and used by INCOSE with permission 11