AI-Based Optimization of Snow Removal Assets / Resources

1
AI-Based Optimization of Snow Removal Assets /
Resources

• Advisory Committee
• Meeting
• April 13, 2001

2
Meeting Agenda

• 200 p.m. Coffee/Refreshments
• 215 Welcome/Introductions -- Prof. Salim
• 225 Progress Report -- Prof. Salim
• 240 Software Issues -- Prof. Salim
• 250 Software Demo Mr. Villavicencio
• Mr. Zora
• 320 Panel Discussion -- All
• 350 Setting of Action Items -- Prof. Salim
• 400 Scheduling of Next Meeting

3
Welcome/Introductions Prof. Salim

• Grant Participants
• Profs. Salim and Timmerman,
• Industrial Technology
• Profs. Emch and Strauss
• Geography

4
Welcome/Introductions Prof. Salim

• Advisory Board Members

Dave Nelson Area Maintenance Manager DOT
-Waterloo Steve Decker Street Department City
of Waterloo Daron Van Helden Public Gov't
Relations AAA Minnesota / Iowa
Budd Curttright Public Works Director City of
Cedar Falls Barbara Berquam GIS Coordinator
Black Hawk County Dennis Burkheimer
Maintenance Safety Services DOT-Ames
5
Project Progress Prof. Salim

• Technical Progress
• Integration of Art Enterprise and GIS package.
• Creation of Databases for resources.
• Software development of a prototype system.
• Demonstration case for Cedar Falls.

6
Project Progress Prof. Salim

• Educational Progress
• Training of graduate students.
• Undergraduate research experiences.
• Project public website www.uni.edu/transport.
• New website exclusive to project under
  development.

7
Project Progress Prof. Salim

• Publicity Goals
• The following article will be orally presented at
  the indicated meeting in Helsinki, Finland, June
  6, 2001.
•  
• Title A Communication-Based Paradigm for
  Construction Asset Management
• Authors MD Salim and Marc A. Timmerman
• Accepted for the proceedings of the 2nd
  Worldwide ECCE Symposium Information and
  Communication Technology (ICT) in the Practice of
  Building and Civil Engineering.

8
Project Progress Prof. Salim

• Submitted for review for oral presentation and
  conference proceeding publication at The Eighth
  East Asia-Pacific Conference on Structural
  Engineering and Construction in Singapore in
  December of 2001
• Title Optimization of Construction Logistics
  Using
• AI-based Heuristic Asset Management Tools
• Authors MD Salim and Marc A. Timmerman

9
Project Progress Summary Table
0

3.6

Final manual

spring 2002/spring 2002

0

4

Dissemination



4.1

Publication and presentation

spring 2002/spring 2002

30

3.2

Website implementation

spring 2002/spring 2002

0

4.3

Revi
sed manual

spring 2002/spring 2002

0

4.4

Revised software

spring 2002/spring 2002

0

4.5

Commercialization

spring 2002/spring 2002

0


10
Review of Project Software
ARTSCRIPT (LISP)
User Interface Input
DATABASE GIS
Problem Set-Up
DATABASE ASSETS
Optimizer
ART ENTERPRISE
ArcView 3.20
No
OK?
Yes
User Interface Output
ArcView 3.20
11
Software Issues Prof. Salim

• Have just obtained GIS network integrator. This
  will be used
• for networking snowplow routes.
• In process of developing better lab network
  interface with UNI
• Computer support staff.
• In process of developing needed CD burn-in
  facilities and
• software for final project distribution.

12
Software Demonstration Mr. Villavicencio Mr.
Zora
13
Snow Removal Asset/Resource Management System

• General prototype. Focus on integration of AI
  shell with GIS and databases
• Use current information of DOT available through
  WWW
• Simple rules for
• Priority assignment of segments of roads, based
  on interstate condition, truck route, high or low
  traffic volume (based on Aadt of 5000 and 2000
  respectively)
• Assignment of snow plower based on cost and
  capacity criteria for each priority
• Requires a hierarchy of rules to provide correct
  results (non monotonic logic)

14

• May be easily extended to include new knowledge,
  restrictions, and other assets
• Can include Cases for learning on CBR engine
• Uses 6 rules, 350 objects, 24 functions/routines,
  8 databases for input, and generates 2 databases

15
Prototype DevelopmentKnowledge Modeling
16
Development of the Prototype

• Implemented using Rapid Development Cycle method
  adapted for AI steps.

• 2 different programming languages
• ArtScript from Art Enterprise (Lisp-like).
  Implements logic and part of database analysis
• Avenue from ArcView (VB-like). Graphic
  representation and data preparation for Expert
  System shell

17
Development of PrototypeArchitecture of the
System

• General View
• Users Perspective

18

• General View
• Developers perspective

19
Output generated by the system

• Initial map generated for Cedar Falls

20

• Priority Assignment and results

21
Panel Discussion -All
22
Wrap Up Prof. Salim

• Setting of action items.
• Plan for next meeting.