A Reuse- and Prototyping- based Approach for the Specification of - - PowerPoint PPT Presentation

▶
a reuse and prototyping based approach for the
SMART_READER_LITE
LIVE PREVIEW

A Reuse- and Prototyping- based Approach for the Specification of - - PowerPoint PPT Presentation

Mar 17, 2024 259 likes •921 views

A Reuse- and Prototyping- based Approach for the Specification of Building Automation Systems Andreas Metzger Stefan Queins SFB 501: Development of Large Systems with Generic Methods Department of Computer Science University of

slide-1
SLIDE 1

A Reuse- and Prototyping- based Approach for the Specification of Building Automation Systems

Andreas Metzger Stefan Queins

SFB 501: ‘Development of Large Systems with Generic Methods’ Department of Computer Science University of Kaiserslautern Germany

slide-2
SLIDE 2

Introduction Requirements Engineering Method

Reuse Prototyping

Case Study Perspectives

A Reuse- and Prototyping- based Approach for the Specification of Building Automation Systems

Andreas Metzger Stefan Queins

SFB 501: ‘Development of Large Systems with Generic Methods’ Department of Computer Science University of Kaiserslautern Germany

slide-3
SLIDE 3

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 3

Properties of Building Automation Systems Control of Physical Effects

temperature

  • ccupancy

sound air quality ventilation light humidity

slide-4
SLIDE 4

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 4

Properties of Building Automation Systems Coupling of Physical Effects

slide-5
SLIDE 5

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 5

Properties of Building Automation Systems Coupling of Physical Effects

slide-6
SLIDE 6

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 6

Properties of Building Automation Systems Coupling of Physical Effects

slide-7
SLIDE 7

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 7

Properties of Building Automation Systems Size

handling of complexity

type of building total sensors and actuators university building with 100

  • ffices and computer labs

2,500 Burj Al Arab and Jumeirah Beach Hotel, Dubai 26,000

slide-8
SLIDE 8

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 8

Properties of Building Automation Systems Size

handling of complexity

Number of Identical Products

small efficient development

type of building total sensors and actuators university building with 100

  • ffices and computer labs

2,500 Burj Al Arab and Jumeirah Beach Hotel, Dubai 26,000

slide-9
SLIDE 9

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 9

Properties of Building Automation Systems Size

handling of complexity

Number of Identical Products

small efficient development

Life Span

long traceability

type of building total sensors and actuators university building with 100

  • ffices and computer labs

2,500 Burj Al Arab and Jumeirah Beach Hotel, Dubai 26,000

slide-10
SLIDE 10

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 10

Development Process Requirements Engineering

Problem Description Requirements Specification

slide-11
SLIDE 11

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 11

Development Process Requirements Engineering

Building Description

Problem Description Requirements Specification

Needs

slide-12
SLIDE 12

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 12

Development Process Requirements Engineering

Building Description Object Structure

Problem Description Requirements Specification

Object Structure Specification Needs

slide-13
SLIDE 13

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 13

Development Process Requirements Engineering

Building Description Object Structure Tasks

Problem Description Requirements Specification

Object Structure Specification Task Description Needs

slide-14
SLIDE 14

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 14

Development Process Requirements Engineering

Building Description Object Structure Tasks

Problem Description Requirements Specification

Object Structure Specification Task Description Needs Informal Object Type Requirements Description

slide-15
SLIDE 15

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 15

Development Process Requirements Engineering

Building Description Formal Object Type Object Structure Tasks

Problem Description Requirements Specification

Object Structure Specification Task Description Needs Requirements Modeling Informal Object Type Requirements Description

slide-16
SLIDE 16

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 16

Development Process Requirements Engineering

Building Description Formal Object Type Object Structure Tasks

Problem Description Requirements Specification

Object Structure Specification Task Description Needs Requirements Modeling Informal Object Type Requirements Description

Domain Dictionary

slide-17
SLIDE 17

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 17

Development Process Development Step

Output Input Activity

slide-18
SLIDE 18

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 18

Development Process Development Step

Output Input Activity

slide-19
SLIDE 19

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 19

Development Process Development Step

Library Output Input Activity

slide-20
SLIDE 20

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 20

Development Process Development Step

Library Defects Check Output Input Activity

slide-21
SLIDE 21

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 21

Development Process Development Step

Library Defects Check Output Input Defects Activity

slide-22
SLIDE 22

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 22

Reuse Expected Benefits

  • high quality products
  • short development time
slide-23
SLIDE 23

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 23

Reuse Expected Benefits

  • high quality products
  • short development time

Reuse Operations

Artifact Creation Candidate Adaption Candidate Specification Effort Estimation Candidate Selection

slide-24
SLIDE 24

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 24

Reuse Expected Benefits

  • high quality products
  • short development time

Reuse Operations Evaluation

  • metric to quantify gain of productivity

–a posteriori assessment to gather experience –a priori estimation to support development decisions

Artifact Creation Candidate Adaption Candidate Specification Effort Estimation Candidate Selection

slide-25
SLIDE 25

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 25

Prototyping Expected Benefits

  • high quality products
  • short development time
slide-26
SLIDE 26

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 26

Prototyping Expected Benefits

  • high quality products
  • short development time

Prototyping Environment

Prototype Generation Formal Model

slide-27
SLIDE 27

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 27

Prototyping Expected Benefits

  • high quality products
  • short development time

Prototyping Environment

Building Simulator Physical Building Interlinking Prototype Generation Formal Model Environment

slide-28
SLIDE 28

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 28

Prototyping Expected Benefits

  • high quality products
  • short development time

Prototyping Environment Evaluation

  • quality

– number and types of errors found

  • effort

– prototype application – benefit for design

Building Simulator Physical Building Interlinking Prototype Generation Formal Model Environment

slide-29
SLIDE 29

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 29

Case Study

O435 O433 O425 CL426 O424 CL422 O416 O423 O415 P413 O412 CL410 CL411 hw1 CL418 O419 O429 hw1 hw1 O414 P420 M427 O431 O421 O417 N

building description:

22 rooms, 3 hallway sections

slide-30
SLIDE 30

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 30

Case Study

O435 O433 O425 CL426 O424 CL422 O416 O423 O415 P413 O412 CL410 CL411 hw1 CL418 O419 O429 hw1 hw1 O414 P420 M427 O431 O421 O417 N

Floor Section Hallway Room sec(3) hw rm(n)

  • bject structure:

25 object types, 920 instances

Object Structure Specification building description:

22 rooms, 3 hallway sections

slide-31
SLIDE 31

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 31

Case Study

O435 O433 O425 CL426 O424 CL422 O416 O423 O415 P413 O412 CL410 CL411 hw1 CL418 O419 O429 hw1 hw1 O414 P420 M427 O431 O421 O417 N

Floor Section Hallway Room HWLight HWOcc RoomLt RoomTm RoomOcc sec(3) hw rm(n) hwOcc hwl rlt rtm

  • cc
  • bject structure:

25 object types, 920 instances

Object Structure Specification building description:

22 rooms, 3 hallway sections

slide-32
SLIDE 32

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 32

Case Study

O435 O433 O425 CL426 O424 CL422 O416 O423 O415 P413 O412 CL410 CL411 hw1 CL418 O419 O429 hw1 hw1 O414 P420 M427 O431 O421 O417 N

Floor SunDet Section Hallway Room HWLight HWOcc HWDoor MotDet MotDet Contact RoomLt RoomTm RoomOcc Door Desk Light T empSens MotDet Contact Radiator Valve T empSens Contact T askLight Contact PulsGen Dimmer PulseGen PIDCtrl Blind sec(3) hw rm(n) hwOcc hwl imd3 door(2) imd(2) dcc rlt rtm

  • cc

door(m) dsk lle(2) tss

  • ls(6)

imd dcc rad (n) rva wts(3) sll tli sll lpg dim lpg pid sdt bld(n) OutDoorLight

  • bject structure:

25 object types, 920 instances

Object Structure Specification building description:

22 rooms, 3 hallway sections

slide-33
SLIDE 33

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 33

Case Study

“Shortly before a persons enters a hallway section, the lights should be turned on, if neccessary.”

68 needs

slide-34
SLIDE 34

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 34

Case Study

“Shortly before a persons enters a hallway section, the lights should be turned on, if neccessary.” “Compute and propagate the light settings considering daylight.” “Determine occupancy using door.” Task

68 needs 126 tasks

Description

slide-35
SLIDE 35

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 35

Case Study

“Shortly before a persons enters a hallway section, the lights should be turned on, if neccessary.” “Compute and propagate the light settings considering daylight.” “If a door reports the possible entrance of a person (newHW- DoorEntrance), set occupancy (occupied) to true and report it (newHWOcc).” Requirements “Determine occupancy using door.” Task

68 needs 126 tasks 126 strategies

Description Description

slide-36
SLIDE 36

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 36

Case Study

“Shortly before a persons enters a hallway section, the lights should be turned on, if neccessary.” “Compute and propagate the light settings considering daylight.” “If a door reports the possible entrance of a person (newHW- DoorEntrance), set occupancy (occupied) to true and report it (newHWOcc).” Requirements “Determine occupancy using door.” Task Requirements

68 needs 126 tasks 126 strategies 37 SDL object types

unoccupied newHW_ DoorEntrance newHWOcc

  • ccupied

Modeling Description Description

Process HWOcc 1(2)

slide-37
SLIDE 37

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 37

Case Study Evaluation

  • total effort:

26,000 min

2000 4000 6000 8000 10000 12000 14000 Object Structure Specification Requirements + Task Description Requirements Modeling Effort [min] Check

slide-38
SLIDE 38

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 38

Case Study Evaluation

  • total effort:

26,000 min

  • reuse operations:

118

– no artifact found: 37 – external: 67 – internal: 14

2000 4000 6000 8000 10000 12000 14000 Object Structure Specification Requirements + Task Description Requirements Modeling Effort [min] Check

slide-39
SLIDE 39

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 39

Case Study Evaluation

  • total effort:

26,000 min

  • reuse operations:

118

– no artifact found: 37 – external: 67 – internal: 14

  • prototyping sessions: 45

– average duration: 82 min – defects (simulation): 23 – defects (phys. env.): 10 – defects (inspection): 24 – prototype generation: 10 min – code size: 10 MB

2000 4000 6000 8000 10000 12000 14000 Object Structure Specification Requirements + Task Description Requirements Modeling Effort [min] Check

slide-40
SLIDE 40

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 40

Case Study Evaluation

  • total effort:

26,000 min

  • reuse operations:

118

– no artifact found: 37 – external: 67 – internal: 14

  • prototyping sessions: 45

– average duration: 82 min – defects (simulation): 23 – defects (phys. env.): 10 – defects (inspection): 24 – prototype generation: 10 min – code size: 10 MB

  • productivity:

– reference is earlier case study – total gain: 94% – gain caused by reuse: 22%

2000 4000 6000 8000 10000 12000 14000 Object Structure Specification Requirements + Task Description Requirements Modeling Effort [min] Check

slide-41
SLIDE 41

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 41

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric
slide-42
SLIDE 42

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 42

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric

Reuse

  • finding reuse candidates
  • effort estimation
slide-43
SLIDE 43

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 43

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric

Reuse

  • finding reuse candidates
  • effort estimation

Prototyping

  • earlier prototyping

Formal Specification Prototype Generation

slide-44
SLIDE 44

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 44

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric

Reuse

  • finding reuse candidates
  • effort estimation

Prototyping

  • earlier prototyping
  • feedback loop

Test Cases Prototype Generation T est Defect Identification Fault Detection Formal Specification

slide-45
SLIDE 45

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 45

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric

Reuse

  • finding reuse candidates
  • effort estimation

Prototyping

  • earlier prototyping
  • feedback loop
  • mixed simulation
  • design support

Test Cases Prototype Generation T est Defect Identification Fault Detection Formal Specification

slide-46
SLIDE 46

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 46

BLANK!

slide-47
SLIDE 47

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 47

Properties of Building Automation Systems Control and Coupling of Physical Effects

slide-48
SLIDE 48

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 48

Properties of Building Automation Systems Size

handling of complexity

Number of Identical Products

small efficient development

Life Span

long traceability

type of building total sensors and actuators university building with 100

  • ffices and computer labs

2,500 Burj Al Arab and Jumeirah Beach Hotel, Dubai 26,000

slide-49
SLIDE 49

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 49

Development Process Requirements Engineering

Building Description Formal Object Type Object Structure Tasks

Problem Description Requirements Specification

Object Structure Specification Task Description Needs Requirements Modeling Informal Object Type Requirements Description

Domain Dictionary

slide-50
SLIDE 50

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 50

Development Process Development Step

Library Defects Check Output Input Defects Activity

slide-51
SLIDE 51

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 51

Reuse Expected Benefits

  • high quality products
  • short development time

Reuse Operations Evaluation

  • metric to quantify gain of productivity

–a posteriori assessment to gather experience –a priori estimation to support development decisions

Artifact Creation Candidate Adaption Candidate Specification Effort Estimation Candidate Selection

slide-52
SLIDE 52

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 52

Prototyping Expected Benefits

  • high quality products
  • short development time

Prototyping Environment Evaluation

  • quality

– number and types of errors found

  • effort

– prototype application – benefit for design

Building Simulator Physical Building Interlinking Prototype Generation Formal Model Environment

slide-53
SLIDE 53

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 53

Case Study

O435 O433 O425 CL426 O424 CL422 O416 O423 O415 P413 O412 CL410 CL411 hw1 CL418 O419 O429 hw1 hw1 O414 P420 M427 O431 O421 O417 N

Floor SunDet Section Hallway Room HWLight HWOcc HWDoor MotDet MotDet Contact RoomLt RoomTm RoomOcc Door Desk Light T empSens MotDet Contact Radiator Valve T empSens Contact T askLight Contact PulsGen Dimmer PulseGen PIDCtrl Blind sec(3) hw rm(n) hwOcc hwl imd3 door(2) imd(2) dcc rlt rtm

  • cc

door(m) dsk lle(2) tss

  • ls(6)

imd dcc rad (n) rva wts(3) sll tli sll lpg dim lpg pid sdt bld(n) OutDoorLight

  • bject structure:

25 object types, 920 instances

Object Structure Specification building description:

22 rooms, 3 hallway sections

slide-54
SLIDE 54

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 54

Case Study

“Shortly before a persons enters a hallway section, the lights should be turned on, if neccessary.” “Compute and propagate the light settings considering daylight.” “If a door reports the possible entrance of a person (newHW- DoorEntrance), set occupancy (occupied) to true and report it (newHWOcc).” Requirements “Determine occupancy using door.” Task Requirements

68 needs 126 tasks 126 strategies 37 SDL object types

unoccupied newHW_ DoorEntrance newHWOcc

  • ccupied

Modeling Description Description

Process HWOcc 1(2)

slide-55
SLIDE 55

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 55

Case Study Evaluation

  • total effort:

26,000 min

  • reuse operations:

118

– no artifact found: 37 – external: 67 – internal: 14

  • prototyping sessions: 45

– average duration: 82 min – defects (simulation): 23 – defects (phys. env.): 10 – defects (inspection): 24 – prototype generation: 10 min – code size: 10 MB

  • productivity:

– reference is earlier case study – total gain: 94% – gain caused by reuse: 22%

2000 4000 6000 8000 10000 12000 14000 Object Structure Specification Requirements + Task Description Requirements Modeling Effort [min] Check

slide-56
SLIDE 56

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 56

Perspectives Development Process

  • expansion: design and new domain
  • workflow concept
  • quality metric

Reuse

  • finding reuse candidates
  • effort estimation

Prototyping

  • earlier prototyping
  • feedback loop
  • mixed simulation
  • design support

Test Cases Prototype Generation T est Defect Identification Fault Detection Formal Specification

slide-57
SLIDE 57

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 57

BLANK PAGE!

slide-58
SLIDE 58

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 58

Reuse Metric

  • reuse operation
  • relative integration effort
  • relative effort for artifact
  • assumption: complexity correlates to effort
  • relative effort for project
  • relative productivity for project

integrate Ra 1 – Ra reuseable artifact artifact a

Erel,integrate Eadapt Ecreate

  • =

a Erel,a Ra Erel,integrate 1 Ra – ( ) 1 ⋅ + ⋅ = Ka Ea ∼ Erel 1 Ka

a

∑

  • Ka Erel,a

⋅

a

∑

⋅ = Preuse Pno reuse

  • Prel

1 Erel

  • =

=

slide-59
SLIDE 59

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 59

Case Study Data

  • correlation

: 0.87 CS3/1 CS3/2 number of object types 25 37 total complexity [units] 688 1826 total effort [min] 19,000 26,000 effort per complexity unit 27.6 14.2 total relative productivity 1 1.94 range of complexity

[units]

6..58 9..114 range of relative effort per artifact 1 0.44..1.35 relative productivity caused by reuse 1 1.22 Prel Ka Erel,a Prel Ka Ea ∼

slide-60
SLIDE 60

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 60

Case Study Quality

  • distribution of errors and removal effort

20 40 60 80 100 5 10 15 20 25 Phase 1 2 3 4 number of errors effort of removal [m in] prototyping applied

slide-61
SLIDE 61

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 61

Workflow Concept

Inspection PrototypeTest Phase 1 Phase 4

  • Req. Eng. Workflows

Calendar Time

Checking ObjStrSpec ObjStrSpec Activities ObjStrSpec Checking

slide-62
SLIDE 62

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 62

Workflow Concept

Phase 5 Phase n Phase 1 Phase 4 Inspection PrototypeTest ObjStrSpec Checking Phase 1 Phase 4

  • Req. Eng. Workflows

Calendar Time

Checking ObjStrSpec Checking ObjStrSpec ObjStrSpec Activities Design Workflows ObjStrSpec Checking ObjStrSpec Checking

slide-63
SLIDE 63

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 63

Testbed Communication Structure

Area µC µC Area µC µC

Workstation

Area µC µC

(Embedded) PC (Embedded) PC (Embedded) PC Ethernet CAN

slide-64
SLIDE 64

Andreas Metzger – OMER-2 Workshop – Herrsching – May 10, 2001 64

BLANK PAGE!