Producing Prosperity - An Industrial Engineer’s Role in Economic Expansion Dr M C Jothishankar Advanced Manufacturing Technology.

Producing Prosperity - An Industrial

Engineer’s Role in Economic Expansion

Dr M C Jothishankar

Advanced Manufacturing Technology

Defining the “New Economy”

Past 100 years, vitality of US economy was determined

by success of “traditional” manufacturing industries -

automobiles, steel, oil, and chemicals

Today, information technology, communications and

intellectual capital determine success

The driving forces of the new economy are ideas,

knowledge, services and higher-order skills

Manufacturing remains important - innovation,

adaptation, and reengineering are the watchwords of

success for today’s workers and businesses.

What Does the New Economy Mean?

To those in information technologies - describes the

power of new communication tools

To venture capitalists - hundreds of investment

opportunities each day

To corporate leaders - new alliances, partnerships or

mergers

To trade advocates - accelarates an integrated, global

economy

To educators - lifelong learning opportunities

To the average citizen - numerous opportunities at home

and at work, and more connectivity worldwide

What are the Characteristics

Products are increasingly non-tangible: knowledge

is the major input

Productivity is increasing: deployment of

technology driving force

Markets are global and competitive: labor and

expertise vs. location and physical structure

Entrepreneurs are spurring economic growth

New partnerships are the wave: co-competition

creates a flexible economy

How Can an IE Become More Competitive in the New Economy?

IEs must build upon their core strengths and focus on the economic foundations of the New Economy:

– Cross functional skills

– Access to new technologies on which new products and processes are based

• Consortium participation

– Collaborative work among:

• Industries, academia, government and labor

Case Study I

Traditional Approach

(Part 1)

Material Movement &Fleet Management

This project aims to study the best possible

routing/distribution of mails/material

between Rockwell facilities in Cedar

Rapids.

Facts Over 30 buildings are served

Seven drivers moved material between buildings

Some stops were delivered 12 times a day

Type of material moved:

– Internal mail

– Dispatches

– “Hot” Dispatches

– Security Dispatches

– Skids

Present Routes

Traditional IE Approach

Time study on the routes

Foot prints of the routes

Cost calculations on resources

Recommendations

Reduce current seven drivers to four drivers

Establish a “hub” at 120 Mailroom for relay of

dispatches between routes

Reduce frequency of visits to a maximum of 8

per day

Final

Results

Total Savings out of this project:

$121,000

Case Study I

Same Problem in New Economy - A Collaborative Approach

(Part 2)

U. S. Region

• Dr. M. C. Jothishankar

• Dr. Dennis Bricker

THE UNIVERSITY OF IOWA

Japan Region

• Mr. Tomomitsu Murano,

• Mr. Hisaya Watanabe • Dr. Seiichi Kawata

European Region

• Prof. Dr.-Ing. Heinz Wörn• Mr. Daniel Frey

University of Karlsruhe, Germany

Quote to Note

“The improvements achieved in one

company can be easily be wasted in the

subsequent phases of logistics chain.”

- Heard on the Street!

Objective of GALAXI

To develop an optimization and simulation

model that will minimize the overall fleet

operations cost and most effectively

distribute material between different

manufacturing plants.

Description of the Model

In this manufacturing system, there are a number

of factories

Each factory manufactures parts, which are used

for assembly of products at the same or another

factory

Parts are transferred between factories by using

several kinds of vehicles (trucks) on demand

Example of Plants

P1

P2

P3

P4

P5

P6

P7 P8

Proposed Solution Method

The problem will be modeled as a

minimum-cost, multi-period, multi-

commodity network flow problem.

One set of variables will specify the

routes and schedules for the trucks,

while another set of variables will

specify the movement of the parts.

Total Cost to be Minimized

cost of vehicles

shippingcosts

storage costs & penaltiesfor late delivery

+ h iktZ i

kti I

t = 1

T

k K

fvmYvmm M

+ ckp lX lkt

k

lt = 1

T

Constraints

capacity restrictions

conservation of flow for each material

limit on # vehicles of each type

integrality of vehicles

Solution Approach

Benders’ Decomposition

Lagrangian Relaxation

Cross-Decomposition

Genetic Algorithm

Simulation

Cross Decomposition

LagrangianSubproblem

Benders'Subproblem

Truck schedules

Lagrangian multipliers

(Determines Material movement)(Solves Lagrangian Relaxation)

Genetic Algorithms

Manual Input

Inter-Relationship Among Models

Cross-Decomposition Method

Genetic Algorithms

Simulation Model

(Y)

Deterministic Model

Stochastic Model

USA

Japan

Germany

Sample Simulation

Expectations of This Project

We hope to reduce our total material movement

cost by 30% a savings of almost $ 250,000

annually　

This software will help the truck schedulers to

make better decisions and to reduce the time

spent in scheduling

Increased truck utilization

Case Study II

Setup Reduction

(Part 1)

Concepts

Look beyond the problem under study -

Instead of “Point” solution approach the

problem to provide a “System Solution”

Use Re-engineering principles

Involve the users

Problem Overview PCB assembly machines have high pick-and-

place rates, but their set-up times are typically

very long

PCBs scheduled in Process Center on first-

come-first-serve basis

Set-up is changed for every PCB batch

Large set-up times and underutilized resources

Setup Details

Kits run per day : 30

Feeder changes between kits : 40

Feeder changes per day : 1200

Time to change a feeder : 30 Seconds

Time to change feeders / month :

1200 x 22 x 0.5 = 220 hours

(10 hours a day!)

Process Center OperationImprovement Objectives

Set-up time reduction

Scheduling time reduction

Increase machine utilization

Decrease manufacturing lead time

Increase throughput

Process Center Optimization Project Approach

PCB manufacturing process reengineering

Development of optimization algorithms

Software development

Simulation studies

Process Center Optimization

Cluster PCBs into groups

Sequence the PCB groups to minimize the total

set-up time

Optimize assignment of feeder locations to

minimize the number of feeder changeovers

Use simulation to evaluate system performance

for generated schedules

Clustering PCBs

Clustering PCBs into a minimum set of groups

such that:

– Groups are formed based on similar components

– Total number of unique component types should

be less than the number of feeders

– Within each PCB group, no set-up is necessary

when changing from one PCB type to another

Clustering PCBs - Example (Before)

Printed Circuit Board

Component

1 2 3 4 5 6 7 8 9

a 1 1b 1 1 1c 1 1d 1 1 1 1e 1 1 1 1f 1 1g 1 1 1 1 1h 1 1

Clustering PCBs - Example (After)

PCB Groups:(4,1,7)(9,8)(2,5)(6,3)

Printed Circuit Board

Component

4 1 7 9 8 2 5 6 3

g 1 1 1 1 1f 1 1a 1 1b 1 1 1d 1 1 1 1e 1 1 1 1c 1 1h 1 1

Optel Procedure

Balance workload on assembly lines

Balance workload on assembly lines

Split assembly components between machines

Split assembly components between machines

Group similar assemblies into families

Group similar assemblies into families

Generate family and assembly sequences

Generate family and assembly sequences

Select work orders to be scheduled for productionSelect work orders to be scheduled for production

Generate machine placement programsGenerate machine

placement programs

Interacting Elements

Production

Plan

PartInventory

Process Center

Data

Schedule

PCB DesignData

O P T E LO P T E L

Case Study II

Manufacturing Optimization

and Execution System (MOES)

for a PCB Assembly Plant

(Part 2)

Present Optel Framework

Plant Data Manager

ERP/MRP

Setup Verification

Material Management

Machine Optimization

Assembly DataModeling

Plant Process Monitor

ShopFloor

OPTEL

Production Scheduling

PDM

Benefits of Using Optel

Setup time reduction: 70%

Increase in machine utilization: 20%

Increase in component placement: 60%

Reduction in machine programming time: 95%

Eliminated night shift and week end operations

Results

Total Savings out of this OPTEL project:

$1.5 M/Year/site

Case Study III

E- Manufacturing

Partners of this E-Manufacturing Project

Virtual Factory

ANY PRODUCT

ANY TIME

ANY WHERE

VisionOur vision is E-manufacturing where we have

Seamless, scalable and robust evolution of

products from design to manufacturing

Computer tools (such as simulators, rule-bases,

visualizing environments) to rapidly plan, validate

and deploy manufacturing instructions

Flexible manufacturing systems for simultaneous

production of multiple products and minimum

system change over

Present flow

Physical Constraintsand Functions

Detailed Design

MFG. Review(DFM Checks)

Design Release

Trial&ErrorProduction

Proposed Flow

ProductionDesign Release

VirtualEnvironment

Physical Constraintsand Functions

EngineeringDesign

Design For Manufacturing

Computer Integrated Manufacturing

Optel - Manufacturing Execution System

Enterprise Resource Planning

Virtual Environment

Process Planners

Simulator

Manufacturing Execution Systems

DesignsRules Resources

ProcessPlans Virtual

ProductsMachinePrograms

ManufacturingAnalysis

Visualizers

RuleInferenceEngines

Flow Diagram

DFM Rule Checker

DFM Rule files

AP 210 file (3D)

AP 210 file (2D)

Assembly configuration

file

Placement Sequence (OPTEL)

Visualizer(STEP OIV)

Machine library

Simulator

Design Facts

2D to 3D converter ECAD translators generate AP 210 files

containing 2D Geometry

Simulate 3D view of the assembly board

Converter

– Input : AP 210 file (2D geometry)

– Extrudes them into solids (Advanced BREP)

– Output:

• AP 210 file (2D + 3D geometry)

• AP 203 files

2D to 3D converter

AP 203 files

– Individual packages

– Board

These files then converted

into Open Inventor format

– Inventor: Graphics

package used for rendering

DFM Rule Facility

Subset of Rockwell Collins DFM rules were

chosen for implementation.

Interface to the simulator

– DFM rules which were violated

– Components which violated these rules.

Assembly Data Establish an assembly usage view

– Components

• Organization by package / part family

• Part numbers, version

• Configuration management data

• Location, orientation

• Reference Designators

Integrated with OPTEL

– Magazine Setup

– Optimized placement sequence

Simulator

Feeder information

Component information

Placement Sequence

Geometric Models

Placement Simulator

Virtual Board

DFM RuleCheck

Components violating

DFM rules

Work Done

Accepting an AP210 2D design file

Manufacturability analysis

Extracting component information

Generating 3D models of components and

assemblies

Generate

– “as designed” view

– “as simulated” view

Rules of New Economy

Change happens

– Anticipate change

Be ready to change quickly and enjoy the

change

– Adapt to change quickly

“It is not the strongest of the

species that survive, not the most

intelligent, but the one most

responsive to change”

Charles Darwin