Slides - Oklahoma State University

Download Report

Transcript Slides - Oklahoma State University

Mixed Model
Production lines
Mixed Model Production Lines
C.S.Kanagaraj
( Kana + Garage )
IEM 5303
Slide 1
2000 C.S.Kanagaraj
Overview
Mixed Model
Production lines
• Mixed Model Production Lines (MMPL)
• Production Planning Techniques
• Evolution of Production Planning
Techniques
• Emergence of a new technique
• Comparisons - Based on the dynamic
environment
Slide 2
2000 C.S.Kanagaraj
Overview (2)
Mixed Model
Production lines
• Problem Statement
• Automated Sequencing System (ASS)
• Steps followed in sequencing
• Conclusion
Slide 3
2000 C.S.Kanagaraj
Introduction
Mixed Model
Production lines
• Mixed Model Production Line (MMPL)
– Production lines capable of making
several different parts for a given period
of time are called as the mixed model
production lines.
• Example: An automobile assembly line
• Mixed Model Production Lines use
various production planning techniques
to enhance their production capabilities
Slide 4
2000 C.S.Kanagaraj
Production Planning Techniques
Mixed Model
Production lines
• What are production planning techniques
– These are a set techniques involving
mathematical equations and algorithms,
which are to used to help us effectively in
scheduling, sequencing, batching, etc. in
the manufacturing industry.
• Example: Genetic Algorithms.
Slide 5
2000 C.S.Kanagaraj
Evolution of Production Planning
Techniques
• 70’s - 80’s
• 80’s
• late 80’s
Optimization Era
Heuristic Era
Artificial Intelligence Era
• 90’s
Interactive Schedulers Era [1]
–
–
–
–
-
Mixed Model
Production lines
Expert Systems
Neural Networks
Genetic Algorithms
Autonomous Agent Architectures
_____________________________________
[1] Maria Caridi and Andrea Sianesi “Multi-agent systems in production planning
and control: An application to the scheduling of mixed-model assembly
lines”, International Journal of Production Economics, Volume 68, Issue 1, 30
October 2000, Pages 29-42
Slide 6
2000 C.S.Kanagaraj
Synoptic Table of Scheduling
Techniques
Mixed Model
Production lines
Era
Control
Approach
Technique
Optimization
Hierarchical
Automatic
Heuristic
Hierarchical
Automatic
Optimization or
Heuristic
Heuristic
Artificial Intelligence
Neural Networks
Genetic Algorithms
Autonomous Agents
Hierarchical
Hierarchical
Hierarchical
Hierarchical
Automatic
Automatic
Automatic
Automatic
Heuristic
Heuristic
Heuristic
Heuristic
Interactive
Schedulers
Distributed
Interactive
Heuristic +
operator
Complexity
[1]
Slide 7
2000 C.S.Kanagaraj
Limitations of these techniques
Mixed Model
Production lines
• Optimization Era
– Differences in mono and multi objective approaches
– long time to reach full automation
– less product differentiation
• Heuristic Era
– More static than being dynamic
– It seems quite impossible to codify the system
reactions to possible failures and events
Slide 8
[1]
2000 C.S.Kanagaraj
Limitations of these techniques (2)
Mixed Model
Production lines
• Artificial Intelligence Era
• Seems to be natural but complex
– Expert System
– Neural Networks
– Genetic Algorithms
• complexity of the software
• difficulty of knowledge codifying
– Autonomous Agent Architectures
• very few are in use at present
• its efficiency and effectiveness are still hard to
evaluate
Slide 9
[1]
2000 C.S.Kanagaraj
Emerging Technique
Mixed Model
Production lines
• Interactive Schedulers Era
– They are the simplest kind of scheduling system
because the plan is not made by a machine but the
planner himself, while the system checks for the
feasibility of the decision makers choices.
– Schedulers are the ones most commonly used
instruments in real world applications.
[1]
Slide
10
2000 C.S.Kanagaraj
Graph - Production Planning
Techniques
Mixed Model
Production lines
Slide
Comparisons - High and Low Dynamics [1]
11
2000 C.S.Kanagaraj
Problem Statement
Mixed Model
Production lines
• A production line of a sheet metal
manufacturing facility is taken into
consideration.
• The problem concerning the production line is
studied. And the area is identified in a transfer
area between the press floor and the surface
treatment zone.
• The parts are sequenced manually before
moving into the next process.
Slide
12
2000 C.S.Kanagaraj
Problem Statement (2)
Mixed Model
Production lines
• To accomplish this task, a basic design and an
algorithm were borrowed from Choi Wonjoon
and Shin Hyunoh [2] and the model was
modified to provide a solution to the above
manufacturing facility.
• The algorithm developed by Choi Wonjoon and
Shin Hyunoh [2] suported a dynamic
environment.
_____________________________________
[2] Choi Wonjoon and Shin Hyunoh “A Real-Time Sequence Control System for the
Level Production of the Automobile Assembly Line”, Computers & Industrial
Engineering, Volume 33, Issues 3-4, December 1997, Pages 769-772
Slide
13
2000 C.S.Kanagaraj
Automated Sequencing System (ASS)
Mixed Model
Production lines
• An automated sequencing system is developed
similar to a PBS system used in by Choi
Wonjoon and Shin Hyunoh [2].
• The automated sequencing system, sequences
the parts according some constraints, like
demand and overall time taken by the part for
getting processed.
• The main aim of the system is to make use of
the production line intelligently and efficiently.
Slide
14
2000 C.S.Kanagaraj
Steps followed for sequencing
Mixed Model
Production lines
1) Selecting the production/feeder line
2) Assigning priorities to the parts according to the
demand
3) Selecting the part with the least processing time
to process next
4) Inserting the selected part to the priority queue
A set of algorithms use by Choi Wonjoon and Shin
Hyunoh [2] are used to perform the sequencing
task.
Slide
15
2000 C.S.Kanagaraj
Conclusion
Mixed Model
Production lines
• Mixed model production lines have
become a universal standard in many
manufacturing facilities. And therefore
when a dynamic sequencing system is
used, the production line becomes highly
productive and efficient.
~ * ~ Best of luck! Guys ~ * ~
Slide
16
2000 C.S.Kanagaraj