Modeling and Analysis of Manufacturing Systems

1
Modeling and Analysis of Manufacturing Systems

• Session 3
• Simulation Models
• January 2001

1
2
Definition of Simulation

• Simulation is the imitation of the operation of a
  real world system over time.
• Simulation involves the generation of an
  artificial history of the system and the drawing
  of inferences from it.

2
3
A First Simulation Example

• One teller bank
• Customers arrive between 1 and 10 minutes apart
  with uniform probability.
• Teller service times are between 1 and 6 minutes
  with uniform probability.
• Goal Simulate the banks operation until 20
  customers are served.

3
4
Questions

• Input data?
• Model vs Reality?
• Length of run?
• Amount of runs?
• Output analysis?

4
5
Modeling Concepts

• System The real thing!
• Model A representation of the system.
• Event An occurrence which changes the state of
  the system.
• Discrete vs Continuous Event Models.
• Dynamic vs. Static Models.

5
6
Modeling Concepts - contd

• System state variables All information required
  to characterize the system.
• Entity An object in the simulation.
• Attributes Entity characteristics.
• Resources A servicing entity.
• Lists and list processing Queues.
• Activities and delays.

6
7
Modeling Structures

• Process-Interaction Method
• Event-Scheduling Method
• Activity Scanning
• Three-Phase Method

7
8
Advantages of Simulation

• Decision aid.
• Time stretching/contraction capability.
• Cause-effect relations
• Exploration of possibilities.
• Diagnosing of problems.
• Identification of constraints.
• Visualization of plans.

8
9
Advantages of Simulation -contd.

• Building consensus.
• Preparing for change.
• Cost effective investment.
• Training aid capability.
• Specification of requirements.

9
10
Disadvantages of Simulation

• Training required.
• Interpretation of results required.
• Time consuming/expensive.
• Inappropriately used.

10
11
Application Areas

• Manufacturing/ Materials Handling
• Public and Health Systems
• Military
• Natural Resource Management
• Transportation
• Computer Systems Performance
• Communications

11
12
Steps in Simulation Modeling

• Problem Formulation
• Goal Setting
• Model Conceptualization
• Data Collection
• Model Translation
• Verification and Validation
• Experimental Design

12
13
Steps in Simulation -contd.

• Production Runs and Analysis
• Documentation/Reporting
• Implementation

13
14
Input Data Representation

• Random Numbers and Random Variates
• X (1/?) ln( 1- R)
• Independent Variables
• Deterministic, or
• Fit a probability distribution, or
• Use empirical distribution

14
15
Verification

• Is the computer implementation of the conceptual
  model correct?
• Procedures
• Structured programming
• Self-document
• Peer-review
• Consistency in input and output data
• Use of IRC and animation

15
16
Validation

• Can the conceptual model be substituted, at least
  approximately for the real system?
• Procedures
• Standing to criticism/Peer review (Turing)
• Sensitivity analysis
• Extreme-condition testing
• Validation of Assumptions
• Consistency checks

16
17
Validation -contd.

• Validating Input-Output transformations
• Validating using historical input data

17
18
Experimentation and Output Analysis

• Performance measures
• Statistical Confidence
• Run Length
• Terminating and non-terminating systems.
• Warm-up period.

18
19
System Dynamics andSimulation Basics
20
System Dynamics

• System
• Collection of Interacting Elements working
  towards a Goal
• System Elements
• Entities
• Activities
• Resources
• Controls

21
System Dynamics (contd.)

• System Complexity
• Interdependencies
• Variability
• System Performance Metrics
• Flow (Cycle) Time
• Utilization
• Value-added Time and Waiting Time
• Flow Rate
• Inventory/Queue Levels
• Yield

22
System Dynamics (contd.)

• System Variables
• Decision Variables (Input Factors)
• Response Variables (Output Variables)
• State Variables
• System Optimization
• Finding the best combination of decision
  variables that minimizes/maximizes an objective
  function

23
System Dynamics (contd)

• Systems Engineering The application of science
  and engineering to transform a need into a system
  with the following process
• Requirements definition
• Functional analysis
• Synthesis
• Optimization
• Design
• Test
• Evaluation

24
System Dynamics (contd.)

• Systems Analysis Techniques
• Simulation
• Hand Calculations
• Spreadsheets
• Operations Research Methods
• Linear and Dynamic Programming
• Queueing Theory (see Harrell p. 42-43)

25
Simulation Basics
26
Simulation Basics

• Types of Simulation
• Static/ Dynamic
• Stochastic/Deterministic
• Discrete Event/Continuous
• Simulating Random Behavior
• Random Number Generation
• Random Variate Generation
• Probability Expressions and Distributions

27
Simulation Basics (contd.)

• Workings of Discrete Event Simulation
• Process Oriented World View
• Sequence of Activities on Entities
• Clock Advancement
• Events Scheduled and Conditional

28
Simulation Basics

• Example
• Single-server queue
• Arrival times uniformly distributed between 0.4
  and 2 minutes. Mean arrival time 1.2 minutes
• Service time 1 minute
• Two Events Arrival and Service completed
• Simulation Table

29
Discrete Event Simulation

• Modeling of a system as it evolves over time by a
  representation in which the state variables
  change instantaneously and only at separate
  (countable) points in time.
• An EVENT is an instantaneous occurrence that may
  change the state of the system.

30
Next-Event Simulation Clock Advancement

• Clock initialized to zero
• Schedule of future events determined
• Clock advanced to the time of occurrence of the
  most-imminent event
• System state updated
• Time of occurrence of future events updated
• Repeat until reaching termination event

31
Components of a DES model

• System state
• Simulation clock
• Event list
• Statistical counters
• Initialization routine
• Timing routine
• Event routine
• Library routine
• Report generator
• Main

32
Simulation Software

• Classification of Simulation Software
• General-Purpose
• Application-Oriented
• Modeling Approaches
• Event-scheduling approach
• Process approach

33
Simulation Software (contd)

• Common Modeling Elements
• Entities
• Attributes
• Resources
• Queues

34
Simulation Software (contd)

• Desirable Software Features
• Modeling flexibility and ease of use
• Hardware and software constraints
• Animation
• Statistical features
• Customer support and documentation
• Output reports and plots

35
DES of a Single Server Queue

• M/M/1 queue
• Mean interarrival time 1 minute
• Mean service time 0.5 minutes
• Find
• Average time in queue? In system?
• Average number in queue? In system
• Server utilization?
• Littles formula?

36
Getting Started
37
Simulation Procedure

• Step 1 Define objective, scope, requirements
• Step 2 Collect and analyze system data
• Step 3 Build model
• Step 4 Validate Model
• Step 5 Conduct experiments
• Step 6 Present results
• Note Iterations required among steps

38
Definition of Objective

• Performance analysis
• Capacity analysis
• Configuration comparisons
• Optimization
• Sensitivity analysis
• Visualization

39
Definition of Scope

• Breadth (model scope)
• Depth (level of detail)
• Data gathering responsibilities
• Planning the experimentation
• Required format of results

40
Definition of Requirements

• The 90-10 rule
• Size of project (data readily available)
• small (2-4 weeks)
• large (2-4 months)
• Data gathering (50 of time)
• Model building (20 of time)

41
The Simulation Project
42
Simulation Project Steps

• a.- Problem Definition
• b.- Statement of Objectives
• c.- Model Formulation and Planning
• d.- Model Development and Data Collection
• e.- Verification
• f.- Validation
• g.-Experimentation
• h.- Analysis of Results
• i.- Reporting and Implementation

43
Basic Principles of Modeling

• To conceptualize a model use
• System knowledge
• Engineering judgement
• Model-building tools
• Remodel as needed
• Regard modeling as an evolutionary process

44
Manufacturing Systems Simulation
45
Manufacturing Systems

• Material Flow Systems
• Assembly lines and Transfer lines
• Flow shops and Job shops
• Flexible Manufacturing Systems and Group
  Technology
• Supporting Components
• Setup and sequencing
• Handling systems
• Warehousing

46
Characteristics ofManufacturing Systems

• Physical layout
• Labor
• Equipment
• Maintenance
• Work centers
• Product
• Production Schedules

• Production Control
• Supplies
• Storage
• Packing and Shipping

47
Modeling Material Handling Systems

• Up to 85 of the time of an item on the
  manufacturing floor is spent in material
  handling.
• Subsystems
• Conveyors
• Transporters
• Storage Systems

48
Goals and Performance Measures

• Some relevant questions
• How a new/modified system will work?
• Will throughput be met?
• What is the response time?
• How resilient is the system?
• How is congestion resolved?
• What staffing is required?
• What is the system capacity?

49
Goals of Manufacturing Modeling

• Manufacturing Systems
• Identify problem areas
• Quantify system performance
• Supporting Systems
• Effects of changes in order profiles
• Truck/trailer queueing
• Effectiveness of materials handling
• Recovery from surges

50
Performance Measuresin Manufacturing Modeling

• Throughput under average and peak loads
• Utilization of resources, labor and machines
• Bottlenecks
• Queueing
• WIP storage needs
• Staffing requirements
• Effectiveness of scheduling and control

51
Some Key Modeling Issues

• Alternatives for Modeling Downtimes and Failures
• Ignore them
• Do not model directly but adjust processing time
  accordingly
• Use constant values for failure and repair times
• Use statistical distributions

52
Key Modeling Issues -contd

• Time to failure
• By wall clock time
• By busy time
• By number of cycles
• By number of widgets
• Time to repair
• As a pure time delay
• As wait time for a resource

53
Key Modeling Issues -contd

• What to do with an item in the machine when
  machine downtime occurs?
• Scrap
• Rework
• Resume processing after downtime
• Complete processing before downtime

54
Example

• Single server resource with processing time
  exponential (mean 7.5 minutes)
• Interarrival time also exponential (mean 10
  minutes)
• Time to failure, exponential (mean100 min)
• Repair time, exponential (mean 50 min)

55
Example 5.1 -contd

• Queue lengths for various cases
• Breakdowns ignored
• Service time increased to 8 min
• Everything random
• Random processing, deterministic breakdowns
• Everything deterministic
• Deterministic processing, random breakdowns

56
Trace Driven Models

• Models driven by actual historical data
• Examples
• Actual orders for a sample of days
• Actual product mix, quantities and sequencing
• Actual time to failure and downtimes
• Actual truck arrival times

57
A sampler of manufacturing models from WSC98

• Automotive
• Final assembly conveyor systems
• Mercedes-Benz AAV Production Facility
• Machine controls for frame turnover system

58
A sampler of manufacturing models from WSC98
-contd

• Assembly
• Operational capacity planning daily labor
  assignment in a customer-driven line at Ericsson
• Optimal design of a final engine drop assembly
  station
• Worker simulation

59
A sampler of manufacturing models from WSC98
-contd

• Scheduling
• Batch loading and scheduling in heat treat
  furnace operations
• Schedule evaluation in coffee manufacture
• Manufacturing cell design

60
A sampler of manufacturing models from WSC98
-contd

• Semiconductor Manufacturing
• Generic models of automated material handling
  systems at PRI Automation
• Cycle time reduction schemes at Siemens
• Bottleneck analysis and theory of constraints at
  Advanced Micro Devices
• Work in process evolution after a breakdown
• Targeted cycle time reduction and capital
  planning process at Seagate

61
A sampler of manufacturing models from WSC98
-contd

• Semiconductor Manufacturing - contd
• Local modeling of trouble spots in a Siemens
  production facility
• Validation and verification in a photolithography
  process model at Cirent
• Environmental issues in filament winding
  composite manufacture
• Order sequencing

62
A sampler of manufacturing models from WSC98
-contd

• Materials Handling
• Controlled conveyor network with merging
  configuration at Seagate
• Warehouse design at Intel
• Transfer from warehouse to packing with Rapistan
  control system
• Optimization of maintenance policies

63
Manufacturing Simulators

• ProModel
• Witness
• Taylor II
• AutoMod
• Arena
• ModSim and Simprocess

• SimSource
• Deneb
• Valisys (Tecnomatix)
• Open Virtual Factory
• EON
• Simul8