How to Improve Process Performance. or Automating Waste Isn t Fun

Size: px

Start display at page:

Download "How to Improve Process Performance. or Automating Waste Isn t Fun"

Nigel Jackson
6 years ago
Views:

1 How to Improve Process Performance Using Lean Automation or Automating Waste Isn t Fun Frank Garcia ADVENT DESIGN CORPORATION

2 What We ll Cover? Building a Lean base for Automation including cellular manufacturing How much Automation is appropriate? p Should it be done in phases or all at once. Review examples of Levels of Lean Automation. Determine how to integrate Data Collection g into an Automation project.

3 AUTOMATE or NOT TO AUTOMATE Need to lower costs & reduce cycle times? Scrap too high? Can t deliver on time? Equipment too slow? Costs too high?

4 X Process Process Process Step1 I Step 2 I Step 3 C/T = 20 min C/0 = 30 min Reject = 5% C/T = 40 min C/0 = 2 hours Reject = 10% C/T = 25 min C/0 = 30 min Reject = 5% I New Machine Process Step 2 I C/T = 5 min C/0 = 10 min Reject = <1% Actual C/T = 20 min

5 What Do We Do? Analyze & Evaluate Using Lean Concepts & Techniques Implement Lean Six Sigma Solutions Design & implement LEAN AUTOMATION and Data Collection!

6 Why Lean Automation? After implementing lean improvements such as cellular manufacturing and setup reduction, selective automation can add value and reduce human variability. ab Richard Schonberger

7 Lean Manufacturing Fundamental Principle of Lean Manufacturing Any activity it or action which h does not add value to the product is a form of waste and must be eliminated or minimized. i i

8 Definition of Value -Added Value is added any time the product is physically changed towards what the customer is intending to purchase. Value is also added when a service is provided for which the customer is willing to pay (i.e. design, engineering, etc.). If we are not adding value, we are adding cost or waste. 90% of lead time is non-value added!

9 Lean Manufacturing Tools Value Stream Map * Flow: Setup Reduction *, Cellular Manufacturing *, Batch Size Reduction, Visual Workplace *,, Layout * Pull: Kanban Systems, Supply Chain Management, Point of Use * Others: Quality Improvement & Analysis *, Total Productive Maintenance, Cross Training * * Used * most frequently prior to Automation

10 Value Stream Map An Assessment Tool The value stream map follows the production path from beginning to end and shows a visual representation of every process in the material and information flows Shows how the shop floor currently operates Foundation for the future state

11 Value Stream Map Concept Suppliers Orders Production Control Orders Customers Schedules I Raw Materials Process Equipment Cycle Times Change Over Reliability Error Rate I Finished Goods Lead Time File: VSM-A1

12 Value Stream Map (Current State) Orders Every 2 Weeks Andrea Aromatics (Scented Oils) New Jersey Porcelain (Round Stones) Alanx (Shaped Stones) Order as Needed Production Control Randomly Placed Orders (Various Sizes) Various Customers 30 Cans of Oil 59,000 Stones 50,000+ Stones Every 2 Weeks Every 2 Weeks Every 2 Months (via stringer) Bi- Weekly Productio n Schedule Daily Shipping Orders Average of 6,000 Stones per Day in Various Size Orders (8 to 20 case & 200 to 400 case range mainly) Daily Shipments Existing Work Cell Soak & Dry Packaging Labeling Cartoning Case Packing Shipping 125 Cans of Oil 20,640 Round Stones 49,000 Shaped Stones Multiple Batch Tanks Ameripack Flow Packager APAI Automatic Stapler I I I I I I up to Operator Operator 1/2 Operator 1/2 Operator 90,504 1 Operator 250 Stones Stones stones in WIP Manual Manual C/T = min. C/T = 1 sec. C/T = 3 sec. C/T = 2 sec. C/T = 1 sec. C/O = 10 min. C/O = 5 min. C/O = 2 min. C/O = N/A C/O = N/A Rel. = 100% Rel. = 85% Rel. = 80% Rel. = 100% Rel. = 100% 11.6 Days 0.7 Days 15.1 days 27.4 Days Lead Time 65 min. 7 seconds 65 minutes, 7 seconds Value-Added Time

13 Value Stream Map (Future State) Orders Every Week Andrea Aromatics (Scented Oils) New Jersey Porcelain (Round Stones) Alanx (Shaped Stones) Monthly Order Production Control Randomly Placed Orders (Various Sizes) Various Customers 12 to 16 Cans 30,000 Stones 25,000 Stones of Oil Once a Once a Week Once a Month Week (via stringer) Bi-Weekly Production Schedule (large orders) Daily Shipping Orders Average of 6,000 Stones per Day in Various Size Orders (8 to 20 case & 200 to 400 case range mainly) Daily Shipments 4 Cases Existing Work Cell Soak & Dry Packaging Labeling Cartoning Case Packing Shipping 75 Cans of Oil 40,000 Round Stones 25,000 Shaped Stones Multiple Batch Tanks Ameripack Flow Packager APAI Automatic Stapler Manual I I I I I up to Operator Operator 1/2 Operator 1/2 Operator 30,000 1 Operator 250 Stones Stones stones in a supermarket in WIP type arrangement with stocking levels by shape and scent Manual C/T = min. C/T = 1 sec. C/T = 3 sec. C/T = 2 sec. C/T = 1 sec. C/O = 10 min. C/O = 5 min. C/O = 2 min. C/O = N/A C/O = N/A Rel. = 100% Rel. = 85% Rel. = 80% Rel. = 100% Rel. = 100% 10.8 Days 0.7 Days Increase Reliability 5.0 days 16.5 Days Lead Time 65 min. 7 seconds 65 minutes, 7 seconds Value-Added Time

14 VSM IMPLEMENTAION FUTURE STATE IMPLEMENTATION PLAN & SCHEDULE LOOP OBJECTIVES PROJECTS SEQUENCE (PRIORITY) COMPLETION DATE (Mon./Yr) Supplier Loop * Develop pull system with suppliers 1. Review Weekly/Monthly Requirements With Suppliers 1 Sep-01 * Increase number of deliveries per week 2. Revise current blanket orders 1 Sep-01 * Reduce raw material inventory to match 3. Send daily consumption data to suppliers 2 Nov Setup point of use raw material areas 1 Oct-01 Production Control Loop * Implement daily shipping schedule 1. Single point to schedule( shipping). See pacemaker loop. 1 Oct-01 * Work to schedule manufacturing based on shipping requirements 2. Daily production schedule by Operations Manager 1 Oct Implement kanban loops as shown on future state VSM 2 Nov Use MRP for materials forecasting 1 Oct-01 Element * Reduce lead time 1. Eliminate WIP between winding & annealing 1 Sep-01 Loop * Develop continuous flow 2. Implement supermarket for element prep and kanbans 1 Oct-01 * Develop pull system with element prep supermarket 3. Crosstrain element assembly operations 2 Dec-01 Module Assembly Loop * Reduce cycle time 1. Analyze process variance in environmental test 1 Oct-01 * Reduce variation in cycle time in testing 2. Analyze process variance in module drydown 1 Oct-01 * Reduce variation in cycle time in drydown 3. Implement supermarket and kanbans for cell installation 1 Oct-01 * Establish pull system with supermarket from cell installation 4. Crosstrain module assembly operations 2 Dec-01 * Reduce use of carriers 5. Analyze need for carriers vs. subassembly units 1 Sep Analyze material handling reductions and layout improvements for element prep operations 3 Jan Set up leak tested subassembly units in supermarket for final assembly 1 Oct-01 Pacemaker Loop * Reduce cycle time 1. Improve assembly methods in final mechancial assembly 2 Dec-01 (Unit Assembly/Shipping Loop) * Establish pull system with supermarket from shipping 2. Review design for manufacturing assembly improvements for tubing assembly 3 Feb Consolidate inspection with final mechanical assembly or shipping/packing 2 Nov Set up raw material suppermarket for final assembly 1 Oct Crosstrain unit assembly operations 2 Dec-01 NOTES: 1. Conduct kanban and point of use training for plant personnel in Priotities as follows: #1(Complete in 2 months), #2(Complete in 4 months), #3(Complete in 6 months) 3. Consider use of teams for implementation after training.

15 Implementing Lean Changes My conclusion is that all of us making a lean leap will need to deal with capability issues (TQM/Six Sigma) and with availability issues (TPM) while removing wasted steps and introducing flow and pull in every value stream (TPS). My further conclusion is that there is no right sequence to follow in tackling these problems. Rather it depends on the nature of the product, the nature of the process technology, and the nature of the business. Jim Womack, October 2002

16 How Do We Use Lean Techniques for Automation? Assess the operation using a Value Stream Map and/or PFDs (Product families & Production data) Evaluate the layout Identify lean improvements & kaizens without automation at o Implement lean improvements using VSM plan Identify lean automation opportunities Design and implement lean automation Start the cycle again!

17 Sigma Level for Sustaining Control 5-6σ: Six Sigma product and/or process design eliminates an error condition OR an automated system monitors the process and automatically adjust critical X s to correct settings without human intervention 4-5σ: Automated mechanism shuts down the process and prevents further operation until a required action is performed 3-5σ: Mistake proofing prevents a product/service from passing onto the next step 3-4σ: SPC on X s with the special causes are identified and acted upon by fully trained operators and staff who adhere to the rules 2-4σ: SPC on Y s BEST 1-3σ: Development of SOPs and process audits 0-1σ: Training and awareness WORST

18 The Lean Automation Cycle Assessment (VSM) Continuous Improvement DO IT! Recommended Solutions Implementation Plan Set Up Layout Cells Visual Automation Information Systems

19 Levels of Automation Load Machine Unload Transfer Levels Machine Cycle Machine Part 1 Operator Operator Operator Operator 2 Operator AUTO Operator Operator 3 Operator AUTO AUTO Operator The Great Divide 4 AUTO AUTO AUTO Operator 5 AUTO AUTO AUTO AUTO As defined by the Lean Enterprise Institute in Creating Continuous Flow

layout Ergonomic problems Extensive material staging No space

20 Level 1 Lean Automation Electrical ca Device Assembly Client wanted wave soldering and robotic pick and place Functional operational layout Ergonomic problems Extensive material staging No space Initially, 13 people in Aurora cell Low output: 300 units/day Reject rate 5 to 8%

21 Lean Six Sigma Techniques Used Process mapping Cellular Manufacturing & Layout Balance Cycle Times Between Work Stations Reduce Batch Size & parts staging Quality Reject Data Collection & Analysis

22 Cellular Assembly Layout Later changed to U Shaped Cells

23 1 WORKSTATION CYCLE TIME: 25sec., 1.25 min. PER 3 UNITS REJECT DATA Cell Changes 2 5to8% TEST SAMPLES Root Cause Analysis LED SOLDER & CUT LED PLACEMENT ASSEMBLY #1 COLD STAKE TEST PCBs CONTACTS ASSEMBLY & SOLDER BUTTON & BATTERY ASSEMBLY LABEL 7 ATTACH BACK COVER, STAKE STRAP & ATTACH STRAP 8 9 GLUE SWITCH/ ATTACH STRAP PACK Pareto Analysis TEST PCBs LED SOLDER & CUT REJECT DATA INSERT SWITCH ACTIVATOR 1 3 ASSIST AFTER CHANGES WORKSTATION CYCLE TIME: 25sec., 1.25 min. PER 3 UNITS REJECT DATA <0.1% PCBs from supplier ASSEMBLY CONTACTS BUTTON & #1 ASSEMBLY BATTERY COLD STAKE & ASSEMBLY TEST PCBs SOLDER LABEL 4 ATTACH BACK COVER, STAKE STRAP & ATTACH STRAP TEST SAMPLES 5 6 GLUE SWITCH/ ATTACH STRAP PACK REJECT DATA INSERT SWITCH ACTIVATOR

24 Lean Six Sigma Changes Cold staking fixtures (Poka Yoke) Powered screw drivers Light test & soldering fixtures Quality data tracking via % defect control chart (p chart)

Better teamwork Faster identification of

25 With Lean Six Sigma Balanced cell at 24 sec per work station Two U-shaped cells 3 piece flow The Results 1000 units/day per cell vs people per cell vs 13 Better teamwork Faster identification of quality problems Operating at 3 to 4 sigma

26 Level 2 Lean Automation Steel Panel Fabrication Wanted to reduce lead time to less than one week Automated equipment had been installed but had problems Long setup times Panel rejects & rework (5%) Few process controls or data collection

27 Lean Six Sigma Techniques Used Process Flow Diagrams Value Stream Mapping Setup time Analysis & Reductions Quality Data Collection & Analysis (Reduce Reject Rate & Variability) Root Cause Analysis

28 Value Stream Map (Current State) Blanket Annual Purchase Order with Daily Releases Takt time = 5 min. Sheet Galvanized Steel (4 by 8 or cut) Sheet Galvanized Steel (4 by 8 or cut) Sheet Galvanized Steel (4 by 8 or cut) Sheet Galvanized Steel (4 by 8 or cut) Production Control (normally working 24 to 48 hours ahead of promised shipment) Randomly Placed Orders (normally single unit orders) Various Distributors (~ 24 for Smith Corp. & ~ 6 for Jones Systems Average volume of 1000 systems per month in peak season. Customers are mainly distributors. There are a few dealers. Up to an average of 130,000 lbs daily in peak season Daily Production Reports Daily Production Reports Daily Shipping Schedule Daily Shipments In Straight Panel Dept. I 2 to 5 days depending on pre-cut size 2 to 5 days Shear Notch Specialty Punch Corner Punch Bend Stake & Label Add Z Brace Radius & Band 1 Accurshear Automated Shear (P-3) 1 Manual Notcher (S-23) & 1 Automated 4 Semi-Auto Punches 3 Semi-Auto Punches (S-1, S-2, & S-3) 1 Manual Brake (R-7) & 1 Automated 1 Automated Machine (R-8) 1 Automated Machine (ACR) 1 Manual Table, 1 Jig-less Machine (R12), Notcher (R-3) Brake (R-13) & 1 Jig Machine 1 Operator 1/2 Operator 0 Operators 1 Operator 1/2 Operator 1/2 Operator 1/2 Operator (R1) 1 Material 2 Operators Handler 2 Operators C/T = 4 min. C/O = N/A Rel. = 99% C/T = 2 min. C/O = 4 min Rel. = 95% C/T = 2 min. C/O = N/A Rel. = 99% C/T = 2 min. C/O = up to 30 min. Rel. = 99% 4 min. C/T = 5 min. C/O = 30 to 60 sec. Rel. = 90% C/T = 2 min. C/O = N/A Rel. = 99% C/T = 7 min. C/O = N/A Rel. = 98% to 99% 2 min. 2 min. 2 min. 5 min. 2 min. 7 min. C/T = 8 min. (average) C/O = 2to30 min. Rej = 5% Uptime = 80% 8 min. Rack C/T = N/A C/O = N/A Rel. = 100% Shipping 2 to 5 Working Days, Lead Time 32 minutes, Value-Added Time

29 INITIAL IMPROVEMENT CONCEPTS Improve reliability and changeover capability of R1 and R12 machines. Reduce panel reject rate. Work to 1 to 2 days lead time Takt time = 5 min. What s Causing Evil Variation? I Radius & Band 1 Manual Table, 1 Jig-less Machine (R12), & 1 Jig Machine (R1) 2 Operators C/T = 8 min. (average) C/O = 2 to 30 min. Reject rate = 5% Uptime = 80% 8 min. Rack 1 Material Handler C/T = N/A C/O = N/A Rel. = 100% 2 to 5 Working Days, Lead Time

30 UNDERSTANDING ROOT CAUSES of R12 PROBLEMS CAUSE AND EFFECT DIAGRAM Red = Most Important Causes SET UP VALUES CHANGE MAINTENANCE CHANGES SETTINGS NO TRUST MOUNTING WRONG DIGITAL READOUT USELESS INDEXES VARY POOR TRANSDUCER SELECTION CRUDE INDEX SYSTEM DESIGN NOTCH O.D.SPACING VARIES.09 IN AIR CYLINDER OPPOSING HYDRAULIC IN HEAD ASSEMBLY EQUIPMENT OPEATOR PREFERENCE USE OF AIR vs. HYDRAULICS INDEXES VARY AIR PRESSURE LOW MAINTENANCE DIFFERENT SETUP PROCEDURES RADIUS VARIES SIDE TO SIDE OPERATORS POOR MAINT PANELS CATCH AT LAST 2 BENDS CONVEYOR NOT ADJUSTED PANEL NOT SQUARE. width TOO LARGE PANEL SQUARENESS OPERATOR JUDGEMENT STRAIGHT PANEL NO SPECS SPECIALTY PUNCH BAD BEARINGS ON LOWER FORMING TOOL RADIUS BACKING SHOE ADJ.USTMENT. PUNCH LOCATION VARIES ANGLES MEASUREMENT NO DIMENSIONAL SPECS OR TOLERANCES SEGMENT LENGTH PANEL WIDTH VARIES BANDS HAVE CAMBER LOCATION PANEL OF SPECIALTY PUNCHES ON PANEL RADIUS TEMPLATE ACCURACY NO SPECS DIFFERENT MEASURES USED ON R1 & R12 3 SUPPLIERS INACCURATE CUTTING PANEL NOTCH POSITION VARIES WIDTH OF STEEL BETWEEN NOTCHES VARIES 3.75 to 4.0 in. ACROSS RADIUS DIFFERENT EQUIPMENT USED NO SPECS MATERIAL (PANELS, STEEL) DIFFERENT STEEL PROPERTIES DON'T MEET CURVATURE TEMPLATE REQUIREMENTS AT SETUP (4' & 6' RADIUS PANELS) SHEET DIMENSIONS VARY GALVINIZED COATING DIFFERENT ON PANELS YIELD STRENGTH VARIES PLATE THICKNESS VARIES NO SPECS SURFACE FINISH VARIES NO SPECS 3 SUPPLIERS HOT VS COLD ROLL 3 SUPPLIERS COATING VARIES 3 SUPPLIERS

31 Process Improvements Lean Based Common setup procedure Replace measurement gages Completed identified R12 maintenance actions Implemented TPM program

32 6 Sigma Based Process Improvements Established radius process capability (3 sigma) DOE for radius variations Implemented process controls for panel dimensions in upstream process Vendor consolidation for dimensional compliance

33 New Radius Bending Machine R13 Automated band cutting Servo driven adjustments from panel bar codes Hydraulic cylinders vs air Online radius measurement and tracking

34 R13 Capabilities After Lean Automation Operates as a cell Runs two product families Changeover in less than 5 sec. within and between product families Cycle time reduced from 5 min. to 1.8 min. Realtime auto check of each panel with data collection Process capability at 4 to 5 sigma

35 Levels of Automation Load Machine Unload Transfer Levels Machine Cycle Machine Part 1 Operator Operator Operator Operator R13 2 Operator AUTO Operator Operator 3 Operator AUTO AUTO Operator The Great Divide 4 AUTO AUTO AUTO Operator 5 AUTO AUTO AUTO AUTO As defined by the Lean Enterprise Institute in Creating Continuous Flow

36 Level 3 Lean Automation Bearing Assembly Functional layout Average batch size of 900 bearings Long set ups of 9 hrs Large amount of WIP Long lead times of 4 to 9 weeks 3 shift operation

37 BEFORE Lean Automation

38 Lean Techniques Used Before Automation Product Family Value Stream Map Set Up Time Reduction (quick changeover chuck) Cellular Manufacturing & Layout Balance Flow & Cycle Time Reduction to Produce Small Batches

39 AFTER Cellular Layout

40 Next Step - Lean Automation Level 3 AUTOMATIC LOADER/UNLOADER Automated parts feeding & reduced handling Decouple machine cycle from operator Cell cycle time at 1 min. per bearing Setup time reduced to 2 to 4 hrs One shift operation Average batch size of 100 down to 10

41 AUTOMATIC LOADER/UNLOADER

42 Across the Great Divide Level 4 Automation

43 Level 4 Automation Smoke Detectors Cellular layout with 7 operators Manual packing Ergonomic problems New product being released Customer wanted to automate the cell

44 AFTER Automation Level 4 Focused product family automation Detectors conveyed between machine operations using special pallets 2 Operators running the line

45 Lean Automation and Data Collection Data, Data, everywhere and nothing do we know. Useful Data becomes Information Useful Information becomes the knowledge base for intelligent decisions and dfuture planning. If knowledge is money then useful data is the currency of the manufacturing floor. As lead time decreases, need for realtime data increases!

46 Lean Automation Can Help Collect the Right Real-time Data What is the purpose of collecting the data? Will the data tell us what we need to know? Will we be able to take action on the collected data?

47 Levels of Control & Information IS DEPT HARDWARE SOFTWARE INFO & CONTROL REPORTING Bus. System MES SCADA PLC Server Farm Large Computers Client/Server PCs Industrial PC Bar Coding PLC & Terminal Large Database ERP/MRP II Large Database Web-based Small Database, SCADA SW PLC Program Fully Integrated Supply Chain Networked to Business System Small Control Network Programmable Control Internet Portal Or Extranet Intranet Portal Ad Hoc Printed Graphs I/O Office PC Spreadsheet Electrical Control By Hand or Printout ENGINEERING

48 Lean Automation Real-time Data to Control Variation for Six Sigma Supports Critical To Quality (CTQ) objectives (3.4ppm defects) Enhances Define-Measure- Analyze-Improve-Control methodology (DMAIC) Process Control Online measurement of process parameters Provide real-time controls as Provide real time controls as limits are understood

49 R13 Process Controls & System Status Real-time Data Collection for Six Sigma Analysis Diagnostics for Rapid Diagnostics for Rapid Identification of Problems

50 How Do Implement Lean Automation? Assess the operation using a Value Stream Map or PFDs (Product families & Production data) Evaluate the layout and flow Identify lean improvements & kaizens without automation Quality improvement TPM Point of Use

51 How Do Implement Lean Automation? Implement lean improvements using VSM plan Identify lean automation opportunities Reduce repetitive motion & stuck at machine Reduce material handling Improve quality Design and implement lean automation & data collection Start the cycle again!

52 Benefits of Lean Automation Lower cost automation Summary Simpler implementation & faster acceptance Greater flexibility for setup & material flow Maximizes operator utilization Better use of floor space Useful data providing a knowledge base for more profitable solutions in the future, and supporting a cycle of Continuous Improvement.

53 Lean Automation Makes Us All Reduced lead times Winners! Reduced costs Shorter cycle times Smaller batch sizes Reduced inventory Improved quality Greater flexibility

54 Contact Information Advent Design Corporation Canal Street and Jefferson Ave. Bristol, PA Frank kg Garcia, Director Planning &P Productivity it

"Value Stream Mapping How does Reliability play a role in making Lean Manufacturing a Success " Presented by Larry Akre May 17, 2007

"Value Stream Mapping How does Reliability play a role in making Lean Manufacturing a Success " Presented by Larry Akre May 17, 2007 LAKRE 2007 1 Lean Manufacturing What is Lean Manufacturing? A philosophy