A Unified Product and Project Lifecycle Model for Systems

Download Report

Transcript A Unified Product and Project Lifecycle Model for Systems

Graduate Students' Perceptions of
Computer-Based Project and
Systems Engineering Management
Methods
Amira Sharon, Olivier L. de Weck, Dov Dori
LINC 2010 MIT
What-When Relations in Systems
Engineering Management (SEM)
ProjectWhen
Management
• Planning
• Organizing
Management Plans
(PMP, RMP…)
Systems
What
Engineering
• Requirement &
Functional
Analysis
Model-Based
Paradigm
Formal and Allocation
Specifications
Design Synthesis
Custom •
Reports
C.1
BMD Command
& Control
*Launch
Command
*Sensor
Commands
Universe
Component
Management Reports
• Directing
and Queries
*Launcher
Status
Launchers
*Interceptor
Status
and Queries
C.2
*Ignition
Collectors
Component
S.1.1
Radar
s1. Inventory
Product
50.1
50.4
s1.Make
Collection
Request
system
s1.
Collection
Request
AND
s1.
F ormatted
Request
50.2
system
AND
AND
50.3
s1.Accept &
F ormat Request
50.5
s1.Get Product
F rom Inventory
s1.Provide
Product to User
s1. Inventory
Product
s1.
Collection
Products
AND
50.6
s1.Check Product
Inventory
user
s1.Accept
Products
s1.
F ormatted
Request
R es ou rc e M I PS
5
0
F unc t ion D et e c t & Init i at e Tra c k
F unc t ion T hreat T rac k
F unc t ion (W it h Ex it s ) D is c rim in a
F unc t ion R eq ues t I nt erc e pt
F unc t ion P erf orm In t erc ep t
F unc t ion A s s es s K il l
0
2
SYS.1
Collection
Management
System
System
• Systems Analysis and
Control (Balance)
• Integration &
• Verification
user
• Control
Customers
Component
Interceptors
*Track Data
• Monitoring
C.3
10
20
30
40
50
57. 0
S.1.2
Analyst/Workstation
Analyst/Command
Center
Component
Component
Project/Product
What-When common
Relationsareas
in
SEM
Adapted from NASA (2005)
Project Management
• Planning
• Organizing
• Directing
• Monitoring
• Control
3
Systems Engineering
Common
Areas
• Risk Mgt.
• Config Mgt.
• Performance
• Evaluation
• Mission
Assurance
• Requirement &
Functional Analysis
and Allocation
• Design Synthesis
• Systems Analysis and
Control (Balance)
• Integration &
• Verification
The managerial part of SEM is conducted
mostly by using traditional PM methods
The Research Context
Conducted within the ProjectProduct Lifecycle Management
(PPLM) research aimed to develop
a combined project-product
methodology and conceptual
model, containing both domains,
linked with explicit relationships.
4
Research Questions 1 & 2
#
Topic
1, Application of
2 Project
Management
Methods within
Systems
Engineering
Management
5
Questions
Research
Method
Notes
(1) While conducting the
systems engineering
management, to what
extent do
practitioners perceive
a notion of a projectdomain, a productdomain, and a
combined projectproduct domain?
(2) How do systems
engineers perceive
the extent to which
the common PM
methods support
SEM?
Structured
focus is put on
questionnaires seven PM
methods
Based on a
UAV case
24 participants
The Unmanned Aerial Vehicle (UAV)
Case
UAV Project Description (NMA-X1)
Situation
You have recently been promoted to Project Manager at New
Millennium Aerospace (NMA) Inc., a leading manufacturer of unmanned
aerial vehicles (UAVs) for the government. Your new job is to plan and
execute the development project for a UAV, to be used for surveillance
purposes. A rough specification and sketch of the new vehicle is shown
in Figure 1. The payload is provided by the government as modified
GFE (government furnished equipment), while the engine will be
supplied by a well-established commercial company (ECC) under a
subcontract. The remainder of the vehicle, including integration and
testing is NMA’s - and therefore your - responsibility. Your task today is
to create a project schedule, find the critical path and to estimate the
finish time of the project. The subsequent project description is
hypothetical, but will help you establish the plan.
The UAV Case – project description story
UAV Project Description (NMA-X1)
The UAV “pusher” vehicle
concept is shown in Figure
1. In a pusher aircraft, the
engine is rear-mounted
which can lead to higher
propulsive efficiency. The
vehicle can be decomposed
into the following
assemblies: fuselage
(houses the avionics suite),
wings, empennage, payload
(a visual and an IR camera,
incl. transmitter) and the
engine (incl. propeller).
After the project start (a,0) you first have to complete the overall requirements definition (b,10) step. Once this
is accomplished you can carry out the following jobs in parallel: negotiate the engine specification (c,5) with
ECC, define your payload specification (d,5), determine the vehicle layout (e,8) and write the software
specification (g,12). You can initiate(GFE) avionics design (f,15) after (b,10), however the tasks (c,5) and (d,5)
must also have been completed before the GFE design can be started, so that the avionics will be able to
control both the engine and payload in a synchronized fashion.
Once the engine specs (c,5) have been defined, the supplier (ECC) informs you that it will take 30 days for
engine development (i,30) based on experience with a previous variant. Once engine development is complete,
delivery and checkout (n,2) can take place at NMA’s facilities. After (d,5) is done, payload development (j,15)
can take place in parallel with engine development. Once the payload is developed (j,15) and the engine
delivered (n,2), both the engine and payload are integrated (electrically) in the power system integration (o,10)
step.
Fuselage design (k,17) and empennage/wing design (l,15) begin in parallel after the vehicle layout (e,8) has
been established. Internal fittings (m,8) can be designed after these two jobs are completed. Also, structural
airframe prototyping (r,8) consists of building a physical frame for the vehicle after jobs (k,17) and (l,15) are
completed.
Once avionics design (f,15) has been completed, this leads to avionics delivery and checkout (p,12) and
subsequent avionics/software integration (q,5). Obviously, in order for this last step to take place, software
development (h,25) which depends both on (g,12) and (f,15) must have also been completed.
The project is continued by performing vehicle integration (s,10) which requires prior completion of power
system integration (o,10), airframe prototyping (r,8) and avionics/software integration (q,5). After vehicle
integration (s,10) and internal fitting design (m,8) have been achieved , final vehicle assembly (t,5) can begin.
After final assembly, the completed vehicle is subjected to laboratory testing (u,5), followed by an outdoor
flight test campaign (v,10) leading to completion of the prototype development project, finish (w,0).
Method for Research Questions 1 & 2
The seven investigated project management methods
Project
management
method –
short name
Project
management
method –
full name
Homework
assignment
Number of 3hour sessions
devoted to
method
Program
Evaluation
System
and
Dynamics
Reviewing
Technique
Critical
Path
Method
Design
Earned
Structure Value
Matrix Method
Object
Process
Methodology
SD
PERT
CPM
DSM
EVM
Gantt
OPM
HW1
HW2
HW2
HW3
HW4
HW5
HW5
1
Reviewed
prior to
HW
Reviewed
prior to HW
6
1
1
2
* The MBPP approach was presented during 1.5 sessions (4.5 hours).
9
Gantt
chart
Instructions to the Participants
The question posed to the participants was phrased as follows:
"Please compare the project models or representations you have
done so far as homeworks, with respect to the following Project
Management (PM) Considerations. Utilize the excel file entitled
HW5 Q4 for this purpose. Wherever you believe a correlation exists
between a model and a PM consideration, provide a short written
explanation of the relationship and grade its strength numerically
(between 1 and 5)."
1 is poor, 2 is fair, 3 is good, 4 is very good, and 5 is excellent. N/A was denoted by 0.
Method for Research Questions 1,2
The 14 Systems Engineering Management Factors
11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
SEM Factor
Budget/Schedule measurement/tracking
Budget/Schedule forecasting
Inter-relationships (process & product)
Resource management
Stakeholders/agents tracking
Performance quality
Product quality
Product planning
Product measurement/tracking
Risk management
Iterations management
Information resolution level
Ease of communication
Change management
Dimension
Project
Project
Project-Product
Project
Project-Product
Product
Product
Product
Product
Project-Product
Project
Project-Product
Project-Product
Project-Product
Method for Research Questions 1,2
Using Alpha Cronbach coefficient
The Alpha Cronbach coefficient α was calculated for estimating how
well the set of variables measures each single one-dimensional
underlying construct.
k
k 
2
2

1   i  T 
k  1  i 1

k is the number of items, i2 is the variance of the ith item, and T is the
variance of the total score formed by summing all the items of the
tested sample.
12
Results for Research Questions 1 & 2
All Factors Set Reliability
Project
Management
Method
Full name
Cronbach's
Alpha
Best Improved
SD
PERT
CPM
DSM
EVM
Gantt
OPM
System
Dynamics
Program
Evaluation
and
Reviewing
Technique
Critical
Path
Method
Design
Structure
Matrix
Earned
Value
Method
Gantt
Chart
Object
Process
Methodology
.743
.793
.754
.640
.757
.760
.855
-
-
-
.702(1)
-
-
-
(1) Improved by deletion of factor 12 – Information Resolution Level and factor 3 - Inter-relationships (process & product)
13
Results for Research Questions 1 & 2
Project Management Methods Comparison
by Sum of SEM Factors Rankings
14
Results for Research Questions 1 & 2
Findings for the defined dimensions
Project
Management
Method
15
SD
PERT
CPM
DSM
EVM
Gantt
OPM
Full name
System
Dynamics
Program
Evaluation
and
Reviewing
Technique
Critical
Path
Method
Design
Structure
Matrix
Earned
Value
Method
Project
Dimension


-

-
-

Product
Dimension

-
-

-
-

Project-Product
Dimension
-
-
-
-

-

Combined
Project-Product
Dimension


-

-
-

Object
Process
Methodology
Results for Research Questions 1 & 2
Project management methods comparison by dimension
16
Research Questions 1 & 2
#
Topic
1, Application of
2 Project
Management
Methods within
Systems
Engineering
Management
Questions
Research
Method
(1) While conducting the
systems engineering
management, do
practitioners perceive a
notion of a projectdomain, a productdomain, and a combined
project-product domain?
Structured
focus is put on
questionnaires seven PM
methods
(2) How do systems
engineers perceive the
extent to which the
common PM methods
support SEM?
17
Notes
24 participants