An Application of Value Stream Mapping to Reduce Lead Time and WIP in a Make-to-Order Manufacturing Line by MASSACHUSETTS INSTITUTE OF TECHNOLOGY Ricolas Wongso NOV 0 4 2010 B.Eng., Materials Engineering (2009) Nanyang Technological University, Singapore LIBRARIES Submitted to the Department of Mechanical Engineering in Partial Fulfillment of the Requirements for the Degree of Master of Engineering in Manufacturing at the Massachusetts Institute of Technology SEPTEMBER 2010 @2010 Massachusetts Institute of Technology All rights reserved A - Signature of Author: Department of Mechanical Engineering August 6, 2010 Certified by: Stephen C.Graves Science Management of Professor J. Siegel Abraham Department of Mechanical Engineering and Engineering Systems Thesis Supervisor Accepted by: David E. Hardt Ralph E.and Eloise F.Cross Professor of Mechanical Engineering Chairman, Department Committee for Graduate Students [This page is intentionally left blank] AN APPLICATION OF VALUE STREAM MAPPING TO REDUCE LEAD TIME AND WIP IN A MAKE-TO-ORDER MANUFACTURING LINE by RICOLAS WONGSO Submitted to the Department of Mechanical Engineering on August 6, 2010 in partial fulfillment of the requirements for Degree of Master of Engineering in Manufacturing ABSTRACT Significant growth in the sales is expected in the coming years for the product family that is the focus of this research. In order to meet the takt time for the future demand, improvement on the current processes and expansion are needed. In this work, Value Stream Mapping was implemented to identify the bottleneck processes: fit up (oval), full welding, and pressure testing. Assembly cell concept, workload balancing and FIFO lanes were proposed countermeasures or improvements to address the capacity shortfall. A decrease of 27% in manufacturing lead time was projected if these improvements were made. In addition, the capacity analysis suggests that an expansion is required in full welding and heat treatment furnace. Keywords: Value Stream Map, Lean, Shared resources, Lead time, Work in Process (WIP) inventory, Make to order, FIFO lane, Capacity expansion, highly customized product Disclaimer: The content of the thesis is modified to protect the identity of the project company. Company name and confidential information are omitted or disguised. Thesis Supervisor: Stephen C.Graves Title: Abraham J.Siegel Professor of Management Science [This page is intentionally left blank] ACKNOWLEDGEMENT First and foremost, I would like to express my deepest appreciation to my thesis advisor, Professor Stephen C. Graves, for his guidance and invaluable insights he provided throughout the duration of the thesis. I am really thankful for his strong commitment to keep in touch despite of his busy schedule, and also for all the time spent during the discussion and video conference. I would also express my appreciation to Dr. Brian W. Anthony who made this group projects possible by making the arrangement with the Company sponsor. In addition, I would like to thank Ms. Jennifer Craig for her guidance and coaching in improving my writing skills. I would also like to convey my thanks to the company that sponsored my work for this thesis. Specifically, I am grateful for my overall corporate supervisor, in providing advice and the necessary resources for me to carry out this study. Many thanks also go to the full time staffs that are involved in the operation of the manufacturing line such as the manufacturing engineers, production planners, quality engineers, etc. for their support in completing this project. Above all, I would like to thank God and my family who have supported me throughout the duration of this thesis and allowed me to focus my attention on completing this undertaking. Last but not least, I would like to thank my team-mates for working together and helping each other during the duration of this group projects. Table of Contents ACKNOW LEDGEM ENT ............................................................................................................................. 5 Chapter 1: Introduction ......................................................................................................................... 10 1.1 Company and Product Description ............................................................................................. 10 1.2 The M anufacturing Process.......................................................................................................... 11 1.3 Current M anufacturing Issues ...................................................................................................... 12 1.4 Thesis Structure .......................................................................................................................... 13 Chapter 2: Problem Statement and Objective ........................................................................................ 14 2.1 Problem Statement...................................................................................................................... 14 2.2 Objective and Scope .................................................................................................................... 16 Chapter 3: Literature Review ................................................................................................................. 17 3.1 W h at is Lean ? .............................................................................................................................. 17 3.2 Introduction to Value Stream M apping (VSM )..............................................................................18 3.3 Application of Lean in Make to Order M anufacturing Environment........................................... 21 3.4 Le a n Glossa ry............................................................................................................................... 23 Chapter 4: M ethodology........................................................................................................................ 24 4.1 Project Road M ap ........................................................................................................................ 24 4 .2 Defin e (D-Phase ).......................................................................................................................... 24 4.3 Measure (M-Phase) ..................................................................................................................... 25 4.3.1 VSM Data Collection.............................................................................................................. 25 4.3.2 Current State VSM ................................................................................................................ 25 4.4 Analyze (A-Phase) ........................................................................................................................ 26 4.4.1 Takt time Analysis ................................................................................................................. 26 4 .4 .2 W IP A nalysis.......................................................................................................................... 27 4.4.3 Capacity Analysis................................................................................................................... 28 4.5 Im p ro ve (I-P hase )......................................................................................................................... 28 Chapter 5: Results and Discussion.......................................................................................................... 29 5 .1 Pro d uct Fa m ily ............................................................................................................................. 29 5.2 Current State VSM ....................................................................................................................... 29 5.2 .1 Past Sal e s Data ..................................................................................................................... 32 5.2.2 Processing Tim e ................................................................................................................... 36 5.2.3 W IP and W aiting Tim e.......................................................................................................... 38 5.2.4 M achine Uptim e .................................................................................................................. 40 5.2.5 First Pass Yield...................................................................................................................... 41 5.2.6 Suppl ier Lead Tim e ............................................................................................................... 42 5.3 Takt Tim e Analysis - Current State ............................................................................................. 42 5.4 Future State VSM ......................................................................................................................... 44 5.4.1 Sizing of FI FO Lanes .............................................................................................................. 47 5.4.2 Sum m ary of Future State VSM .......................................................................................... 49 5.5 Capacity Analysis.......................................................................................................................... 50 Chapter 6: Recom m endation ................................................................................................................. 6.1 Recom m endation on the Current State ..................................................................................... 6.1.1 FIFO Lanes Im plementation................................................................................................. 6.1.2 Production Scheduling.................... ..................... ..................................................... 52 52 52 52 6.1.3 Fit Up, Plasm a Cutting and Pocket Fitting Assem bly Cell .................................................... 53 6.1.4 Heat Treatm ent Im provem ent........................................................................................... 53 6.1.5 W orkload Balancing ............................................................................................................. 54 6.1.6 Other Im provem ents............................................................................................................54 6.2 Recom m endation on Capacity Expansion ................................................................................. 55 Chapter 7: Conclusion and Future W ork ............................................................................................. 56 7.1 Conclusion ................................................................................................................................... 56 7.2 Future W ork................................................................................................................................. 56 Re fe re nce s ............................................................................................................................................ 57 APPENDIX A - VSM Icons ....................................................................................................................... 58 APPENDIX B - Product Fam ily Selections ............................................................................................. 59 APPENDIX C - Current State VSM for Round Product Fam ilies ............................................................... 61 APPENDIX D - VSM Data........................................................................................................................ 67 7 Table of Figures 11 Figure 1: General process flow chart of oval and round products ..................................................... 1 to 24)....................14 Figure 2: Current weekly production output of the assembly line - 2010 (week Figure 3: Exam ple of a current-state M ap [7]..................................................................................... Figure 4: Current state VSM of Oval product family (1/2)................................................................... Figure 5: Current state VSM of Oval product family (2/2)................................................................... 19 30 31 33 Figure 6: Comparison of actual sales vs. forecasted demand ............................................................. 34 ................................................................ Figure 7: Percentage of demand of round product families 34 Figure 8: Percentage of demand of oval product family sorted by size .............................................. 35 Figure 9: Percentage of demand of oval product family sorted by type of materials .......................... 35 Figure 10: Percentage of demand of round product families sorted by size ....................................... Figure 11: Percentage of demand of round product families sorted by type of materials....................36 Figure 12: Takt time and cycle time comparison for both oval and round products - Current state........ 44 45 Figure 13: Future state VSM of Oval product family (1/2) .................................................................. 46 Figure 14: Future state VSM of Oval product family (2/2) .................................................................. Figure 15: Takt time and cycle time comparison for both oval and round products - Future state ......... 49 58 ................. .............................................................................. Figu re 16 : VSM ico ns [7]................ 59 Figure 17: Previous Work of Product Family Selections..................................................................... 60 Figure 18: Current Product Fam ily Selections.................................................................................... 61 Figure 19: Current state VSM of Round-standard product family (1/2) ....................... 62 Figure 20: Current state VSM of Round-standard product family (2/2) .............................................. 63 Figure 21: Current state VSM of Round-guardrail product family (1/2) ....................... 64 Figure 22: Current state VSM of Round-guardrail product family (2/2) .............................................. ..................65 Figure 23: Current state VSM of Round-integral product family (1/2) ......... 66 Figure 24: Current state VSM of Round-integral product family (2/2) ................................................ List of Tables Table Tab le Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 1: Projected number of products to be produced per week (in pieces)..................................... 12 24 2: Project road map ..................................................................................................................... 3: Example of processing time calculation for a product family for specific process or station ...... 26 32 4: Sum mary of current state VSM ............................................................................................ ........ 37 ........ 5: Weighted average processing time for every product family (in hours) . 38 .......................................... hours) (in families product 6: Cycle time of processes for different 39 7: Waiting tim e of Oval products product fam ily ....................................................................... 40 8: Waiting time of Round products product families ................................................................ ........ 41 ......... 9: Machine downtime of 2009 in hours ................................ 41 10: First Pass Yield (FPY) data .................................................................................................. 42 ......................................... process outsourced 11: VSM data on 43 12: Takt tim e of processes in the assem bly line ....................................................................... 48 ............. .................. 13: Calculation of size of FIFO lanes ........... 14: Summary of future state VSM of Oval product family.......................................................... 49 15: Maximum weekly output of the assembly line (after improvement)................................... 50 16: Maximum weekly output vs. forecast weekly demand for 2011, 2012 and 2013 (in pieces) ....51 67 ................................... 17: Processing time of oval products (in hours) 68 18: Processing time of round products (in hours) .................................................................... 19: Number of products queuing before processes (oval products).......................................... 69 20: Number of products queuing before processes (round products)....................................... 70 21: Lead time of outsourced processes quoted by suppliers..................................................... 70 Chapter 1: Introduction 1.1 Company and Product Description Company X is a multi-national company that provides technology, information solutions, and integrated project management services to its customers. Company X has their Engineering, Manufacturing and Sustaining facility for their products in Singapore. The plant is equipped with a foundry, machine shops, assembly shops, a heat treatment furnace and a comprehensive set of Quality Control testing facilities. The products are highly customized and have a relatively low demand (high mix - low volume). They are manufactured in different shapes (oval and round), diameters (2 3/8", 2 /8", 3 1/2", 41/2" and 5 1/2"), lengths, and materials (410SS, 4130, 410-13Cr, Super 13Cr, Inconel 925, etc.), and can be categorized into various product families accordingly. However, significant manufacturing differences only occur between two groups of products; products that have round cross-sections and products that have oval cross-sections. Round products can be further broken down into two types; standard round products and long round products. Standard round products are made up of six components, while long round products are made up of four components. Oval products, on the other hand, consist of four slightly-different main components. -A m- I . . .......... ".." ....... ..... ..... ......... . ........ ...... ...... ...... Moog" ,, 1.2 The Manufacturing Process The general manufacturing process flow for oval and round products is shown in Figure 1 below. ASSEMBLY LINE Figure 1: General process flow chart of oval and round products The overall manufacturing process can be separated into the machining and assembly lines. As shown in Figure 1 above, the four components of the oval products are forged by an outsourced supplier and need not go through the in-house machining line. These forged components go directly into the assembly line which starts with the Fit-Up process and ends with Packaging and Shipping. The four or six components of round products, on the other hand, are first machined in-house from raw bar-stock purchased from an external supplier before proceeding downstream into the assembly line. All assembly workstations that the round and oval products flow through are shared. Both types of products have similar assembly processes and are treated equally. There are four inspection processes in the assembly line that are highlighted in blue, namely Radiography, Hardness Test, Magnetic Particle Inspection (MPI), and Pressure Test. Processes in italic font in Figure 1 are currently outsourced processes, namely Radiography, Threading, Sand-blasting, and Painting. As mentioned in Section 1.1, the products are highly customized in nature and have part numbers as component identifiers. Multiple parts, each with a component part number, are then given a new serial .... .. .. .. .... .... ... . ... ................ ....... ............ number when they are assembled together to identify the particular product. The manufacturing process flow might differ from time to time depending on specific customer requirements. Currently there are more than 50 different designs of oval products and more than 40 different designs of round products. Each time a new product design with new features is released, new product part numbers are created. A more detailed process flow about the different products will be covered in Section 5. 1.3 Current Manufacturing Issues Significant growth in the sales is expected in the coming years. Efforts to increase manufacturing capacity to meet customer demand with competitive lead time, cost, and quality have been put in place. Table 1 below shows the projected number of products that need to be produced per week over the next few years. Table 1: Projected number of products to be produced per week (in pieces) Year 201 Oval 38 Round 9 Tta 2011 77 14 91 2012 2013 103 116 19 23 122 139 47 The current manufacturing capacity is able to produce a maximum of 50 pieces per week. Thus, it will be necessary to improve the throughput rate in order to meet the projected demand over the next few years. Also, capacity expansion projects might become a need should the above trend in future demand materialize. Also, current manufacturing lead times are much longer (ranging from six to nine weeks) than total processing times due to excessive Work-In-Process (WIP). This results in high non-value-adding (waiting) times. It is noted that achieving less WIP will reduce waiting times and consequently reduce the manufacturing lead times. Shorter manufacturing lead times will in turn enable a quicker response to customer orders, ensuring better customer service and on-time delivery. 1.4 Thesis Structure This thesis is organized into chapters. Chapter 1 is an introduction to the company, the product and the manufacturing process. Chapter 2 describes the problem statement, objectives and scope of the thesis. Chapter 3 reviews work on Lean concepts, Value Stream Mapping (VSM), application of lean in make to order manufacturing environment and Lean glossary. Chapter 4 describes the method that the author used to solve the problem. Chapter 5 describes the current state VSM, takt time analysis, future state VSM, and capacity analysis. Chapter 6 describes the recommendation on the current state and capacity expansion. Chapter 7 is conclusion and future work. A.. .. ....... '-w ............ . .... ... .............. Chapter 2: Problem Statement and Objective 2.1 Problem Statement As mention in Section 1.3, there are several issues that need to be addressed with regards to the performance of the manufacturing line: they are the rapidly increasing forecasted customer demand, long manufacturing lead time and high amount of WIP. Figure 2 shows the current production output of the assembly line. Weekly Production Output - 2010 (week 1 to 24) 60 \ Ir..V 30 / '\ \ /0000\ 40 oe V 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Week Figure 2: Current weekly production output of the assembly line - 2010 (week 1 to 24) As can be seen from Figure 2, the current manufacturing line is still capable to support the demand for 2010 which are 47 products per week as shown in Table 1. In order to meet the projected customer demand of 2011, the line capacity needs to be doubled. In order to increase capacity, cycle time of the processes should be reduced so that the products can be produced faster. However, cycle time reduction to increase capacity has its limit. At some point, capacity expansion will be required in order to increase capacity, especially if the customer demand is increasing rapidly. The capacity expansion could be done by buying new machines, tools or equipments, and hiring more workers. The machining line has a relatively small amount of manual work because Computer Numerical Control (CNC) machines ........... are used. In the contrary, the assembly line has large amount of manual work. In addition, processes such as fit up, welding, hot straightening, cold straightening, etc. require highly skilled workers. Currently, the company already has a plan on what kind and how many machines need to be acquired, and also how many workers will be needed to increase the capacity. This plan was created based on prior analysis and experience of the manufacturing engineers. From then, a lot of improvement projects have been done to improve the manufacturing line. Hence, it is important to re-access the current performance of the line and to revise the expansion plan, so that necessary budget reallocation could be done to effectively improve the line. The next issue will be the long manufacturing lead time and high amount of WIP in the manufacturing line. These issues are closely related because manufacturing lead time can be cut by reducing WIP in the system (reduce waiting time) and also by reducing the cycle time of the processes. The current manufacturing lead time of the assembly line for oval and round products is about 5 to 6 weeks. The current WIP in the manufacturing line is estimated to be 180 to 200 pieces combining oval and round products according to the manufacturing engineer. The management feels that there is an opportunity to reduce the manufacturing lead time because most of the lead time is the waiting time between processes. It can be observed that products are waiting to be processed almost before every station. There is an opportunity to address the above issues using Lean manufacturing strategies. However, due to limited resources of time, manpower and money, sometimes it is not feasible to do all improvements simultaneously. Value Stream Mapping (VSM), which originates from Lean manufacturing concepts, is one such tool that can be used to identify and eliminate waste in the manufacturing line. After these areas that need to be improved have been identified, an implementation plan could be developed to make priority to critical areas and quickly solve the problems. This project seeks to improve the performance of the manufacturing line by Lean implementation and also identify areas where capacity expansion is required. 2.2 Objective and Scope The objectives of the project are as the following: * Identify and eliminate waste and bottlenecks in the current manufacturing line * Improve the overall process flow, reduce manufacturing lead time and reduce WIP e Identify critical areas where capacity expansion are needed The scope of this project is the manufacturing line for both oval and round products, focuses more on the assembly line. Chapter 3: Literature Review The following section describes the approaches that are proposed by the author in order to solve the problems. The author will use Lean tools and techniques which are widely used by companies around the world in order to achieve the similar objectives. 3.1 What isLean? Lean manufacturing is a set of principles that continuously identifies and eliminates sources of waste in the entire value chain. The core idea is to maximize customer value while minimizing waste [1]. Lean manufacturing was originated from Toyota Production System (TPS) and identified as lean only in the 1990s [2]. There are 8 types of waste identified in lean manufacturing (can be abbreviated as D-O-W-NT-I-M-E) which are Defects, Over-production, Waiting, Non-engaging employees, Transportation, Inventory, Motion, and Excessive-processing. Defects are defined as bad parts or out of specification parts that need to be reworked or need to be scrapped. Over-production is defined as producing product ahead of demand. Waiting is defined as idling time of parts waiting to be processed, for example, waiting for equipments, operator or raw materials). Non-engaging employees could be defined as poor use of human intellect or work force. Transportation is defined as unnecessary movement of products or materials that is actually not required for processing. Inventory is defined as raw materials, work-in-process (WIP) inventory and finished product that are not being processed. Keeping inventory requires space and there are costs associated with it. Motion is defined as unnecessary motion in the operations, for example, equipment or operator movement. Excessive-processing could be defined as doing non-value added process to the products. There are five Lean Principles which are described in the book, 'Lean Thinking', as shown below [3]. 1. Specify Value. Value is described as what the customer is willing to pay for. One example could be processes which transform the product, e.g. machining, assembly, etc. It is defined only by the customer; however sometimes it could be distorted by pre-existing organizations, especially engineers and experts. They add complexity of no interest to the customer. 2. Identify the Value Stream. The Value Stream is all the actions or processes needed to bring a product or deliver value to the customer. The complete value stream flows through the complete supply chain, from raw materials to finished goods 3. Flow. The value-creating steps or processes should be made to flow without delay or interruption. One should try to eliminate departments that execute a single-task process on large batches. 4. Pull. The production should be make-to-order. The production processes should be activated when the customer wants to receive, not when the supplier wants to provide. 5. Pursue Perfection. There is no end to the process of reducing time, space, cost and mistakes. One should strive for perfection by continually reducing waste. Lean principles have several strengths. They provide a structured methodology for diagnosing and executing waste elimination. Lean focuses on workplace organization and preventative techniques. It is also very effective at rapidly reducing operational costs. Typical results that are obtained after Lean implementation in manufacturing systems include shorter lead times, increased productivity and efficiency, less inventory, lower overall production costs which leads to higher profit and return on assets, cleaner work areas, and waste elimination [4]. Some limitations of Lean principles include that it does not bring a process under statistical control [5]. In other words, it is not capable of removing bottlenecks driven by process variability or defects [4]. Lean relies heavily on intuition, or trial-and-error problem-solving, hence it could be a weakness when a problem is caused by interactive factors and makes problem resolution complex [6]. 3.2 Introduction to Value Stream Mapping (VSM) According to the book "Learning to see" published by the Lean Enterprise Institute (LEI), Value Stream Mapping (VSM) is a pencil and paper tool that helps you to see and understand the flow of material and information as a product makes its way through the value stream [7]. The VSM helps the user to visualize multiple processes and see the flow. It is also useful to create a blue print to implement lean concepts because it allows the user to identify the waste and also source of waste in the value stream [7]. Figure 3 shows an example of a VSM. Information about VSM icons are shown in Figure 16 in Appendix A. ... ............... -It A- _ 46W '4_ Figure 3: Example of a current-state Map [7] In order to get started, first the user needs to choose a product family. A product family is defined as a group of products that pass through similar processing steps and over common equipment in your downstream processes [7]. It is very common that a value stream has more than one product family. In that case, one may consider choosing product families which have higher demand because it is not practical to create the VSM of every single products. It is more beneficial for the user to focus on a product family that has significant value to the customers. The next step would be to visit the plant or the shop floor in order to observe and start mapping the current state of the value stream. It would be helpful to begin with a quick walk along the value stream to get a feel for the flow and sequence of the processes and also interact with operators in the shop floor to learn more about the process. User should pay attention to the material and information flow. User also need to complete the process box data with lean measurement such as process cycle time, changeover time, yield, machine uptime, inventory, etc. Lastly, user would need to draw a timeline and accumulate the process time and inventory waiting time to get the production lead time, value adding time and also non value adding time. ............. After the current state has been completed, the user should asses the current state value stream in terms of creating flow by waste elimination. Then the user should try to draw the future state VSM which will be the ideal state after lean implementation and waste in the value stream has been eliminated or minimized. Several Lean guidelines that are useful to aid the user in developing the future state VSM are as the following [8]: 1. Produce according to takt time It is defined as the customer demand rate. Takt time is used to synchronize the pace of production with the pace of sales to avoid over-production or under-production. Takt time is calculated by dividing the effective working time by the total demand for the products. 2. Finished goods strategy An example of this strategy is the supermarket pull system. Finished goods are being kept as inventory, and then when customers place the orders the finished goods are withdrawn from the supermarket. The upstream processes respond by supplying the supermarket with the same number of goods that were withdrawn (pull system). 3. Develop continuous flow wherever possible Continuous flow is described as producing one piece at a time with each piece being passed to the next process without waiting. 4. FIFO ("first in, first out") lane It isoften used for highly customized parts, parts that have short shelf life or if the parts if costly. It is also useful to control WIP inventory between processes by setting a maximum size of the lane. If the maximum size is reached, the upstream process should stop producing. 5. Pull system Customer process withdraws the items it needs from the inventory and the supplying process produces to replenish what was withdrawn from the inventory. 6. Schedule only at one point The point at which the work is scheduled is called the pacemaker. This pacemaker process will be the one which control how fast the value stream need to operate. 7. Interval It is defined as the time needed to cycle through all the products in the family. Interval is a measure of batch size and flexibility. Smaller interval provides more flexibility to the manufacturer to build what the customers want, when they want it. 8. Pitch It refers to the frequency in which the work are released or taken away from the pacemaker. The future state will help the user to create an implementation plan in order to achieve the ideal state. The implementation plan should consist of improvement activities equipped with a timeline to keep track of the progress. After all the activities have been completed, the user could continuously improve the value stream by repeating the cycle of creating the current state VSM, assess, map the future state and implement improvement. 3.3 Application of Lean in Make to Order Manufacturing Environment Make to order strategy is normally used when a manufacturer needs to produce highly customized product with high mix of different products. The challenges of implementing lean in this manufacturing environment are the difficulties to see the flow because of variations in cycle time of the processes and also because of sharing of equipment or resources. However, some of the basic lean tools such as 5S and visual management could still be applied. 5S stands for Sort, Set in order, Shine, Standardize and Sustain [9]. It is a lean tool which focuses on workplace organization. Sort means only keep necessary tools or equipments which are needed in a workstation. Set in order refers to arranging these tools or equipments in an orderly manner and labeled in order to promote efficient workflow. Shine means keeping the workplace tidy and organized. Standardize means that the first 3S should be consistent and standardized so that all workers know their responsibilities. Sustain refers to maintaining and reviewing the first 4S. Examples of visual management in a manufacturing line are production control board, defect rate, production target, etc. They could be in a form of electronic display boards as well as printed posters. The book "Learning to see" provides some guidelines in implementing Lean in make to order manufacturing environment. First, the scheduling point or the pacemaker process should be further upstream. Second, FIFO lanes could be used to substitute for a supermarket and maintain flow between the downstream processes. For example, supermarket pull system could be used to withdraw the components or raw materials from the warehouse, and then the downstream process are linked by FIFO lanes. In addition, the value stream should keep the interval very small so that it could respond faster to customers demand. The author also reviewed some of the previous works in Lean implementation in make to order manufacturing environment to learn about Lean tools and implementation that are useful or suitable for this type of environment. Making a High-Mix Make-to-Order Production System Lean is a Master Thesis by Bo Li [10]. The objective of the thesis was to improve the manufacturing lead time, reduce WIP and also improve information flow [10]. The method used was to implement CONWIP (Constant Work in Process) system and run a simulation to determine the optimal size for the WIP. CONWIP pull system uses a single global set of cards to control total WIP anywhere in the system [11]. The WIP is not controlled at individual workstation but at a system level, easier to implement and adjust, since only one set of system cards is used to manage system WIP [11]. Kitting is a process in which parts or components needed for manufacturing or assembly of certain products are being collected together into a kit, and issued at the point of use. In addition, the material flow between processes inside the system are maintained and controlled by implementing FIFO lanes. Head & Base Production Optimization: Setup Time Reduction is a Master Thesis by Haiqing Quo [12]. The main objective of the thesis was to reduce the setup time of machine without increasing scrap rate and improves output and productivity of the whole machining section [12]. The method used was SMED (Single Minute Exchange of Die) which is a lean tool to reduce the setup time or changeover time when manufacturing different products. There are eight techniques proposed by Shigeo Shingo, leading manufacturing expert of the Toyota Production System, which can be used to achieve this: to separate internal from external setup operations, to convert internal to external setup, to standardize function, to eliminate fasteners, to use intermediate jigs, to adopt parallel operations, to eliminate adjustments, and mechanization [13]. As results, 45% setup time reduction was achieved. 3.4 Lean Glossary The following list provides the definition of terms that will be used extensively throughout the thesis: * Lead time (L/T): the time required for one piece to move all the way through the process or value stream, from start to finish e Manufacturing lead time: the time required for one piece to move all the way through the manufacturing process; in this project this consists of the machining line and assembly line. e Supplier lead time: the time required for the supplier to deliver raw stocks or forged components, from the time the order is placed until it arrives in the company's warehouse. " Product family: a group of products that pass through similar processing steps and over common equipment in the downstream processes e Value added time (V/A): the time taken for the processes that transform the product in a way for which the customer is willing to pay * Takt time: the customer demand rate * Processing time: the time a product is actually being worked on by an operator or a machine * Cycle time: how frequently an item or product is completed by a process. For example, a welding process takes 4 hours to finish a product; if there are 4 welders available in the station, the cycle time of the welding process is 1 hour. * Bottleneck: a process that cannot meet takt time * Queue time: The time a product spends waiting for the next processing step . ..... Chapter 4: Methodology In the following section, the author will describe his approach to solve the problems based on the Lean DMAIC method that is commonly used in the company in any improvement project. DMAIC stands for Define, Measure, Analyze, Improve and Control. Lean DMAIC consists of five phases and it is structured to help the user to solve the problem systematically. 4.1 Project Road Map In this project, the author only used four phases of this method which are the DMAI phases. Table 2 shows the list of actions planned by the author that will be carried out in each phase. Table 2: Project road map DEFINE MEASURE 0 0 Understand the current manufacturing process Create a problem definition, objective and scope of the project * Analyze existing or historical data * Collect VSM data on the shop floor e Map the current state VSM * Identify bottlenecks in the current manufacturing process Takt time analysis ANALYZE IMPROVE 0 Capacity analysis 0 Map the future state VSM e Identify area for improvement * Identify area for expansion 4.2 Define (D-Phase) In order to understand the operations of the current manufacturing process, the author conducted an informal interview with the people involved in the operations of the manufacturing process, for example the manufacturing engineers, production planners, purchasing, customer service, production manager, supervisor, operators, etc. .... ....... In addition, he went to the shop floor and observed how the operators do their work in each process, and also to learn about the information flow and product flow in the manufacturing line. This is very important to gain an initial understanding about the production line and also to get some clues that might be useful in solving the problems. Next, the author defined the problems, the objective or goals that he istrying to achieve, and the scope of his project which were described in Section 2. 4.3 Measure (M-Phase) 4.3.1 VSM Data Collection The author acquired existing or historical data from the people involved in the manufacturing line to complete the information needed for creating the current state VSM. Examples of such information are historical sales data, forecasted customer demand, general process flow, products routing, manufacturing lead time, first pass yield, WIP and inventory status, cycle time, machine downtime and supplier lead time. These findings are available in Section 5.2. The author decided to conduct a time study at the shop floor to measure the processing time for all the processes. Data on WIP between processes, availability of operators and number of shifts available are also collected and is reported in Section 5.3. 4.3.2 Current State VSM To get started, the author grouped the existing products into product families based on their manufacturing process flow which was extracted from the products routing data. Then, the author analyzed the historical sales data in order to determine the demand percentage for these product families; based on this the author decided to focus more on the product families with higher demand. The forecasted customer demand data for 2010 was used to calculate the weekly demand of the product families. The assumptions that are made for the VSM are as the following: * Available hours per shift = 8.5 hours - 0.5 hours (lunch) - 2x0.25 (tea-breaks) - 0.5 (toolbox meeting) = 7 hours e 3 shifts per day; 5 working days per week; 52 weeks per year .... ........ . ........... . ............ . ...... .-............... ...... ....... ...... .. ........ .. .... . .... In order to determine the processing time of each product family, the author used weighted average method based on the demand percentage of the products within product family. The demand percentage was obtained from the historical sales data and was sorted based on the product's diameter and type of material. The processing time for each product was obtained from time a study on the shop floor. The calculation method using arbitrary numbers is illustrated in Table 3 below. Table 3: Example of processing time calculation for a product family for specific process or station A2 A3 A4 2 1.5 1 Total 40% 10% 25% 35% 0.4 0.2 0.38 0.35 1.33 Finally, after all the data mentioned in Section 4.3.1 have been collected, the author used software named Igrafx [14] to assist him on creating the VSM for four product families. As mentioned in Section 2.2, the author focused more on the assembly line for both oval and round products. On the VSM, the machining line will be simplified by combining together the process steps from the machining cell into a single stage. The reason behind this simplification is because the machining routings for the round components are not fixed, so the same parts could be machined in different ways. 4.4 Analyze (A-Phase) From the current state VSM, the current manufacturing lead time, value added time and WIP could be determined. Next, the author did further analysis as described in the following sections to access the current state of the manufacturing line. 4.4.1 Takt time Analysis The purpose of this analysis is to identify the bottleneck processes in the line and to determine how much improvement is needed so the takt time could be met. Takt time was calculated by dividing the effective working time by the total demand for the products. In practice, factors such as number of shift, operator availability, first pass yield, and machine uptime need to be considered in calculating takt time of a process. Total demand for the products for a process also depends on whether it is a dedicated resource or a shared resource. The formula to calculate takt time for a process step is shown below. Takt time = Effective working time Total demand for the products no. of days per week x no. of shift x avail. hours per shift x operator availabilityx machine uptime weekly demand + first pass yield In a dedicated resource, the weekly demand is the demand for one single product family. However, in a shared resource, the weekly demand is the accumulated demand for several product families that need to be processed in this shared resource. The demand for each round product family could be determined from the demand percentage of these product families. Next, the calculated takt time values were compared to the cycle time for each of the processes. If the cycle time is larger than the takt time, it indicates that the process is a bottleneck. 4.4.2 WIP Analysis The purpose of this analysis is to determine the waiting time of a product between processes. This waiting time (in days) will be displayed in the VSM. The formula to calculate the waiting time is shown below. . number of mandrels bef ore process capacity of a process number of mandrels bef ore process Effective working time per day + cycle time The number of products before each process will be determined from observation at the shop floor. It will be counted at the start of the morning shift for each day for one week. The number of products waiting before the process will also be displayed in the VSM. 4.4.3 Capacity Analysis The purpose of this analysis is to identify processes that would need capacity expansion in order to meet the projected demand for 2011 and beyond. It is done by calculating the maximum weekly output of a process (assuming it works for 3 shifts) and comparing it to the forecast weekly demand. If the weekly output of the process is smaller than the forecasted weekly demand, expansion is needed for this process. The formula to calculate the maximum weekly output isshown below. Maximum weekly out put = Available time per week x machine uptime cletmofhepcssx cycle time of the process yield 4.5 Improve (I-Phase) In this phase, the author mapped the ideal future state of the assembly line after implementing Lean Strategies to the manufacturing line. Examples of the Lean strategies used were FIFO lanes with maximum inventory size and assembly cell concept. Kaizen bursts icons (see Figure 16 in Appendix A) were drawn on the VSM to show the improvement activities that need to be done at a process. Capacity analysis was also done in order to identify areas that need expansion in the future. Finally, recommendations to solve these problems were proposed. Chapter 5: Results and Discussion 5.1 Product Family As mentioned in Section 3.2, to get started, the products must be sorted into product families based on their process steps. Although the products are highly customized, generally they can be grouped based on their shape, size and material. Preliminary work had been done by one of the manufacturing engineers who grouped the products into five product families. This work was passed on to the author, who then analyzed the work, verified the process steps and revised it. Details of the product family selection are shown in Figure 17 and 18 in Appendix B. The author combined group B and group E product families because the process steps are similar for both product families. The product families that have been identified and analyzed in the next following sections are: 1. Oval 2. Round -standard 3. Round - guard rail 4. Round - integral 5.2 Current State VSM The current state VSM shows the process flow from the start to the end of the manufacturing line for each product family. In order to simplify the VSMs for round-products product families, the processing time of machining stage in the round-product families is assumed to be 3 to 4 weeks, which is determined from the company's integrated system (MfgPro) that is used to run the operations._The reason behind this simplification is that the machining cell consists of 5 machines and there are too many different machining routings for the different components. The current state VSM of Oval product family is shown in Figure 4 and 5. The current state VSMs for the round-products product families are available in Appendix C.The VSMs are separated into two sections (upstream and downstream) for better viewing. 29 days 02 days 0.0 days | 1.2 days 0.1days | 0.0 days 0.0days || 0.3 days 0.3 days 0.4 days 0.1days || | Figure 4: Current state VSM of Oval product family (1/2) 30 1.1days 0.8 days | 0.1days || 0.0 days || 0.9 0.6 days Customer Demand: 38 pieces per Week (Takt Time 2.76 hours) 0.9 days 0.6 days 0.1 days 1.0 days 3.1 days 0.8 days 0.6 days 0.0 days 0.0 days 0.0 days 12.0 days Figure 5: Current state VSM of Oval product family (2/2) Manufacturing UT = 26.8 days Total P/T = 13A days Waiting Time = 13.4 days From the VSM, the manufacturing lead time, total processing time and waiting time could be calculated. This information is summarized in Table 4 below. Table 4: Summary of current state VSM Product Family Manufacturing Lead Time (days) Total Processing Time (days) Waiting Time (days) 26.8 13.4 13.4 Roundstandard 55.5 42.5 13 Roundguardrail 57.6 43.7 13.9 Roundintegral 56.3 43.3 13 The current state VSMs indicate that the manufacturing line operates as a make-to-order and push system. After customer order is confirmed, the work order is then released to the shop floor. In the assembly line, the product is pushed to the next station one after another. There is space for inventory in between the processes. More information about the value stream such as processing time, waiting time, number of products between processes, defect, machine uptime, operator availability, and number of shifts are also displayed. The following sections further explain the data and information that were used to construct the VSM. These data and information were extracted from the existing data given from the planners and engineer, and also observed from the shop floor. 5.2.1 Past Sales Data First, past sales data were used to analyze the accuracy of the projected demand. The comparison between the actual sales and projected demand for year 2009 and 2010 is shown in Figure 6. ................... Monthly Sales Data 180 160 140 120 100 80 60 40 20 0 _ , Oval Round - Oval (projected) - Round (projected) Figure 6: Comparison of actual sales vs. forecasted demand Figure 6 suggests that for the first half of the year 2009 and 2010, the actual sales seem to fall below the projected demand. However, for the year 2009, the sales were catching up during the second half of the year. Similar trend might happen for 2010, so the manufacturing line should have enough capacity to meet the projected customer demand. The forecasted demand for 2010 is displayed in the current state VSMs as weekly customer demand which will be used in the takt time analysis in Section 5.3. Second, the author sorts the sales data to get the percentage of demand of the round product families so that the author can focus more on analyzing a product family which has higher demand. The result is shown in Figure 7 below. It indicates that the demand is spread evenly among these three product families so the author decides to include all four product families (including oval product family) in the analysis. ..... ....................... .. .......... ... .... . ....... ......... Percentage of Demand of Round Products (sort by Product Family) 45% 40% 39% 35% 30% 25% 20% 15% 10% 5% 0% Round-standard Round-guard rail Round-integral Figure 7: Percentage of demand of round product families Then, the author further sorts the data to find the percentage of demand of products within product family based on diameter size and type of materials. Figures 8 and 9 show the percentage of demand of oval product family. Figures 10 and 11 show the percentage of demand of round product families. Percentage of Demand of Oval Product Family (sort by Size) 45% 40% 400% 40% 36% 35% 30% 25% 20% 19% 15% 10% 5 5%5 0% 23/8 27/8 31/2 41/2 Figure 8: Percentage of demand of oval product family sorted by size 1-111,11-- I'll, ....... ..... ......... ...... A- OwNWOW YY o __ - __ - Percentage of Demand of Oval Product Family (sort by Material) 39% 45% 40% 35% 30% 25% 20% 15% 10% 410-5 5% 0% LHT 4130 HHT 4130 410SS Figure 9: Percentage of demand of oval product family sorted by type of materials Percentage of Demand of Round Product Families (sort by Size) 100% 90% 86% . 80% 70% 60% * Round-standard 50% " Round-guard rail 40% 1 Round-integral 30% 20% 10% 0% 8% 2 7/8 31/2 41/ 2 5 1/2 Figure 10: Percentage of demand of round product families sorted by size .... ... ... Percentage of Demand of Round Product Families (sort by Material) 100% 90% 80% 70% 60% _92 69% * Round-standard 50% * Round-guard rail -_ 40% 30% 20% 10% * Round-integral 8% 0O% =3 9H8% 0% n 0Z 410-13CR Super 13CR Incornel 925 Figure 11: Percentage of demand of round product families sorted by type of materials The percentage of demand data shown in Figures 8, 9, 10 and 11 are not displayed in the VSM; however they will be used to determine the cycle time data of a product family using weighted average method. It will be further explained in the next section. 5.2.2 Processing Time The results of the time study for oval and round products are shown in Tables 17 and 18 in Appendix C. Some of the processing time for products with certain material and size are not available especially for round products. In that case, the demand percentage is adjusted so that the sum of the weighing factor is still 100%. For example, the processing time for round products with size 2 7/8 and 3 1/2 are not available, so when the round-standard processing time is calculated, the weighting factor for different size of products is assumed to be 100% instead of 86%. Next, the author calculates the weighted average processing time for each product family based on the processing time data on Tables 17 and 18 in Appendix C,and also the percentage of demand data from the previous section. Table 5 shows the results of the weighted average processing time for the product families that will be displayed in the VSM and used in the analysis. ..... ... . ... . ........................ . ... . ... . ... ........ . ........................ 44 V ...... ....... ................... ... ................. ....... . . ..... ... Table 5: Weighted average processing time for every product family (in hours) Process Fit up (Oval) Oval 1 Average Processing Time of Product Family Round-integral Round-Standard Round-Guardrail - - Plasma Cutting 0.8 - Pocket Fitting 0.9 - - ID Welding - 2.7 2.6 2.5 Fit up (Round) - 1.8 1.8 1.7 Mill Slots/Flats - - 5.3 10.4 7 2.1 1.5 11.6 1.2 0.9 0.5 0.75 1 8.1 1.6 1.5 17.3 1.4 1.1 0.5 0.75 1 7.9 1.6 1.4 18.2 1.4 1 0.5 0.75 1 7.5 1.6 1.7 17.1 1.6 1.1 0.5 0.75 1 Full Welding Hot Straightening OD Grinding Heat Treatment Cold Straightening OD Drift Hardness test MPI Pressure Test Mill Slots/flats is a machining process, but it is also part of the assembly process. The setup time and run time done in this machine are recorded in MfgPro. Data is extracted from the MfgPro, and then all the setup and run time to mill slots or flats from January 2010 to June 2010 were summed. Next, these values are divided by the quantity of products being machined to get the average processing time. The processing time is the sum of setup and run time. The processing data is used to calculate the cycle time of the processes by dividing it by the number of stations or batch sizes. The results are shown in Table 6 below. The cycle time data are used in the takt time analysis of the current state VSM. .. .......... ............... ... ........ ...... ... ..... . ...... .... ...... Table 6: Cycle time of processes for different product families (in hours) Oval Processing Cycle time time Round-standard Processing Cycle time time Process No. of stations/ batch sizes Fit up (Oval) 1 stations 1.0 1.0 Plasma 1 stations 0.8 0.8 - - 1 stations 0.9 0.9 - ID Welding 1 stations - - Fit up 1 stations - Slla 1 stations Full Welding Round-guardrail Processing Cycle time time Round-integral Processing Cycle time time - - - - - - - - - - - - 2.7 2.7 2.6 2.6 2.5 2.5 - 1.8 1.8 1.8 1.8 1.7 1.7 - - - - 5.3 5.3 10.4 10.4 3 stations 7.0 2.3 8.1 2.7 7.9 2.6 7.5 2.5 2 stations 2.1 1.1 1.6 0.8 1.6 0.8 1.6 0.8 OD Grinding 1 stations 1.5 1.5 1.5 1.5 1.4 1.4 1.7 1.7 Heat batch sizes 11.6 0.7 17.3 1.1 18.2 1.1 17.1 1.1 1 stations 1.2 1.2 1.4 1.4 1.4 1.4 1.6 1.6 OD Drift 1 stations 0.9 0.9 1.1 1.1 1.0 1.0 1.1 1.1 Hardness 1 stations 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 MPI 1 stations 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 Pressure Test 1 stations 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 - Cutting Pocket Fitting (Round) Mill Slots/Flats Hot Straightening Treatment Cold of 16 Straightening Test 5.2.3 WIP and Waiting Time The number of products waiting to be processed between stations was counted at the start of each morning shift. The challenges faced were that the operators are not dedicated into a particular process and some of the processes were only running in one shift. Hence, the WIP was constantly changing and the long queue was transferred from one process to another. ... ....... ............ ...... .......... . There are two methods to determine the WIP data to be put in the VSM. First is by taking a snapshot or one time observation at the shop floor. Second is by monitoring the WIP status for a period of time and then take the average. The latter method was chosen because by taking more data samples, the validity of the data over a period of time could be assessed. Five and six days' data were collected and the results are shown in Tables 19 and 20 in Appendix C.The averages of these data samples are taken to determine the WIP between processes to be displayed in the VSM and also to calculate waiting time between processes. Next, waiting time of products between processes shown in VSM was calculated using the data provided in Tables 6, 19 and 20. For a shared resource, the number of products waiting before the process is the total of oval and round products. The number of products waiting and the calculated waiting time for oval and round product families are shown in Tables 7 and 8 below. Table 7: Waiting time of Oval products product family Process Fit up (Oval) Plasma Cutting Pocket Fitting Full Welding Hot Straightening OD Grinding MPI Heat Treatment Hardness test Cold Straightening OD Drift Pressure Test No. of units before process 20 2 1 11 5 5 7 18 13 5 Effective working time per day (hour) 6.9 6.9 6.9 20.8 21.0 20.4 7.0 11.9 7.0 9.5 Cycle Time (hour) 1.0 0.8 0.9 2.3 1.1 1.5 0.8 0.7 0.5 1.2 Waiting time (days) 2.9 0.2 0.1 1.2 0.3 0.4 0.8 1.1 0.9 0.6 8 9.5 0.9 0.8 7.0 1.0 3.1 22 ... ...... .. .... ... ... .... ........ .. ...... .. .. .... .. .... .. .......... .......... .... .... ................................................. ..... .... .......... .................... .. .................... .. ..... ........... ..... ............ ... ..... .... . ..................................... ... ..................... ....... . . ........... .... ...... ........ ...... .... ............. ............... ... .. .. .... ... ... Table 8: Waiting time of Round products product families ID Welding Fit up (Round) Full Welding Hot Straightening OD Grinding Heat Treatment Hardness test Cold Straightening OD Drift MPI Pressure Test Mill Slots/Flats No. of units before process 2 0 11 5 5 18 13 5 8 7 22 0 Effective working time per day (hour) 6.65 7 20.79 21 20.37 11.9 7 9.45 9.45 7 7 21 Cycle Time (hour) 2.6 1.8 2.6 0.8 1.5 1.1 0.5 1.5 1.1 0.8 1.0 7.9 Waiting time (days) 0.8 0.0 1.4 0.2 0.4 1.7 0.9 0.8 0.9 0.8 3.1 0.0 The waiting time of the products to be sent to the outsourced processes, such as radiography, threading, sand blasting and painting, is assumed to be 1 day because the products are shipped every 1 or 2 days. Number of products before the station is determined from the supplier data provided in Section 5.2.6. 5.2.4 Machine Uptime Table 9 shows the total downtime duration of the machine in assembly line in 2009. These data were obtained from the maintenance system. These values are converted into machine uptime assuming that the available time in a year is 8640 hours (360 days x 24 hours/day). The reason behind this assumption is that the downtime duration of each machine is the actual real time. And so 8640 hours were used as the available hours even though the actual working hours will be much less. The machine uptime will be displayed in the VSM and used in the takt time analysis. ................... - - .......... . ..... ...... ........ .......... ........ ................ ................. .. .... .... .. .. ..... ...... Table 9: Machine downtime of 2009 in hours 2009 Process Plasma Puasma cutting Jan Feb Mar Apr 1 13 0 0 May Jun Jul Aug Sep Oct Nov Dec 2 0 0 24 14 24 0 1 Total _____ 79 U09ptime 99% _______ ID welding 0 12 0 0 0 100 25 296 0 0 0 0 433 95% Full welding 1 4 15 0 3 80 0 2 2 1 0 0 108 99% OD grinding 18 0 104 0 104 0 46 0 8 4 5 1 290 97% 0 9 8 0 0 2 500 359 175 190 6 4 1253 85% 0 0 0 0 0 0 0 7 0 0 1 0 8 100% 1 1 0 0 0 10 17 0 1 6 2 0 38 100% Heat treatment Cold straightening Kent Milling machine The machines or equipment in the assembly line are pretty stable, except for the heat treatment furnace. The time to repair is very long. When the furnace is down, the company outsources the heat treatment process to a vendor. This will make sure that the other processes still could run. 5.2.5 First Pass Yield Table 10 shows the First Pass Yield (FPY) of the inspection processes. For MPI and radiography, if the product fails the inspection, it will be reworked in the full welding process. For hardness test, it will be reworked in the heat treatment process. The FPY data will be displayed in the VSM and used for takt time, capacity and manpower analysis. Table 10: First Pass Yield (FPY) data FPY % Radiography FPY % Hardness test FPY % Pressure Test FPY % 90.4% 95.7% 68.1% 90.4% 92.3% Feb 2010 85.5% 94.5% 73.9% 91.6% 97.3% Mar 2010 76.1% 98.0% 65.1% 92.2% 99.1% Apr 2010 62.0% 96.8% 71.2% 93.4% 99.4% May 2010 79.0% 100% 70.6% 91.5% 94% Average 78.6% 97% 70% 91.8% 96.4% Machining MPI FPY % Jan 2010 Month -... -- ----------------- ..... .. . 5.2.6 Supplier Lead Time There are three processes that are outsourced to suppliers which are radiography, threading, and sandblasting and painting. Table 21 in Appendix C shows the lead time quoted by the respective suppliers. The lead times depend on the quantity of products being sent to the suppliers. The products are being shipped out almost every one or two days. The company monitors when the products are being shipped out and received from suppliers. Data from period January to May 2010 are analyzed to obtain the actual average lead time and average shipment quantity. The results are shown in Table 11. Table 11: VSM data on outsourced process Average Supplier Lead Time (days) 4 8 4 Radiography Threading Sandblasting and Painting Average Shipment Quantity (pieces) 5 7 10 These data are displayed in the current state VSM as the number of products before the outsourced processes and also the processing time of the outsourced processes. These lead times were assumed to be the processing times of the outsourced processes to simplify the calculation of total manufacturing lead time. 5.3 Takt Time Analysis - Current State Table 12 show the summary of all the information displayed in the current state VSM and also the calculated takt time of each of the processes. The takt time calculation of the fit up (oval) process is shown below as an example. Takt time = Effective working time Total demand for the products no. of days per week x no. of shift x avail. hours per shift x operatoravailability x machine uptime weekly demand + first pass yield 5 x 3 x 7 x 33% x 100% 38 0.9hours/piece .. . . ...... iiiii!-!!iiip9!iiiii!-!!i.-...-... ........... The weekly demand for full welding, heat treatment and hardness test are 50 because for roundguardrail product family, these processes were done twice. Welding of guardrail is included to the full welding because the processing time is similar. Table 12: Takt time of processes in the assembly line Process No of Shift/day Operator Availability Machine Uptime Weekly Demand First Pass Yield Takt Time (hours/piece) 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 85.0% 100.0% 100.0% 91.8% 100.0% 100.0% 91.8% 97.0% 96.4% 0.9 0.9 0.9 3.7 3.9 17.5 1.8 1.3 2.2 1.1 1.6 1.1 0.6 2.4 0.7 (pieces) Fit Up (Oval) Plasma Cutting Pocket Fitting ID Welding Fit Up (Round) Mill Slots/Flats Full Welding Hot Straightening OD Grinding Heat Treatment Cold Straightening OD Drift Hardness Test MPI Pressure Test 3 3 3 2 2 3 3 3 3 2 3 3 1 1 1 33% 33% 33% 50% 50% 100% 100% 60% 100% 100% 70% 50% 100% 100% 100% 100% 99% 100% 95% 100% 100% 99% 100% 97% 85% 100% 100% 100% 100% 100% 38 38 38 9 9 6 50 47 47 50 47 47 50 14 47 Next, the takt time were compared with the cycle time provided in Table 6 to determine whether a process is a bottleneck. If takt time is less than the cycle time, that process is identified as bottleneck process. Figure 12 illustrates the takt time and cycle time comparison. . ... ..... ....... Takt Time vs. Cycle Time (Current State) MTakt Time 8 Oval N Round-standard U Round-guardrail * Round-integral 20.0 18.0 16.0 14.0 12.0 10.0 o 8.0 6.0 4.0 2.0 n .0 00 X A A~ Ao AN .(.00~ A@ A(~ . " 4 Figure 12: Takt time and cycle time comparison for both oval and round products - Current state Figure 12 suggests that the bottleneck processes are: fit up (Oval), full welding, and pressure test. In order to meet customer demand, these bottleneck processes need to be eliminated. 5.4 Future State VSM The future state VSM is only constructed for the oval product family because most of the processes are shared resources, thus any improvements made on these shared resources will also benefit the other product families. The future state VSM for oval product family is shown in Figures 13 and 14. The improvement activities are shown in the VSM by the Kaizen burst icon. A more detailed explanation about these improvement activities are provided in Section 6.1. FIFO lanes are also proposed to replace the traditional push system. The maximum sizes of the FIFO lanes are shown in Figure 13 and 14. The waiting time shown in the future state VSM was calculated from the cycle time and number of products in the FIFO lanes, as described in Section 5.2.3. The FIFO lanes were assumed to be filled up so the number of products is assumed to be equal to the maximum size of FIFO lanes. A more detailed explanation about the sizing of FIFO lanes is provided in the following section. 8 mandrels 1 days 1.0 , 1 d1.0 0.3 d0.3 0 1 days || days 0.3 days || 0.3 days 0.1 days || 0.9 0.0 d1.3 0.1 days || Figure 13: Future state VSM of Oval product family (1/2) 45 0.6 d1.1 0.0 days || d 0.6 days | 0- dys 1.1 days 00ds[1.1 0.6day days 0. das 01dyj0.8 .1 ays | days | 0. wS 0.0das || 1.1 days 0. as 00 dys 0.5 days Figure 14: Future state VSM of Oval product family (2/2) 46 | 9.0 days jTota Manufacuin Lff = 19.8 days P/T = 10.4 days Waiting Time = 9.4 days 5.4.1 Sizing of FIFO Lanes Both oval and round products will enter the FIFO lanes. If the FIFO lane reaches its maximum capacity, the upstream process of the FIFO lane has to stop, and the downstream process has to start to clear the queue. The products that need to be reworked will also enter the FIFO lanes; however, they are being prioritized and are allowed to cut the queue. The maximum sizes of the FIFO lanes were determined by practical consideration and also scenario analysis. The general scenarios are as the following: 1. If the available working time of the upstream process is equal or larger than the downstream process: o Because both processes produce according to the takt time, the cycle time for the upstream process will be equal or larger than the downstream process. Hence, the FIFO lane will be filled, only if the downstream process is idling while the upstream process isstill running. o The FIFO lane should be able to accommodate the products coming from upstream process when the downstream process is idling. When both processes are running, the FIFO lane will decrease because the downstream process is much faster than the upstream process. o The maximum inventory that will pile up between the two processes will be the differences between the processes in available working time per day divided by the cycle time of the upstream process. This inventory will be set as the maximum size of the FIFO lane. 2. If the available working time of the upstream process is smaller than the downstream process: o Because both processes produce according to takt time, the cycle time of the downstream process is larger than the upstream process. Hence, the FIFO lane will reach maximum if the upstream process is running and the downstream process is idling. However, in this scenario, it is unlikely to happen because the available working time of the upstream process is smaller than the downstream process. o When both upstream and downstream processes are running, the inventory between processes will depend on the discrepancies between cycle time of the upstream process (arrival) and the cycle time of the downstream process (departure). o The inventory between processes will be the output of the upstream process subtracted by the output of the downstream process during a period of time when both processes are running simultaneously. This inventory will be set as the maximum size of the FIFO lane. The details of the calculations of the FIFO lanes are shown in Table 13. . .. ........ .... .............. .. . .. ........ ...... .. .. .. .... ...... .................. ...... .. - .......... - .- ... ........ .. . ..... ............ -.-...... -1... 11- - - - l-....... ........................... ............... . . .............. . . ------------..... ........ .... .. .......... ... .. .. . . .... ... .. .... --- ----............. ............ ....... ... ....... ............... --. -.... .. ...... ...... Table 13: Calculation of size of FIFO lanes FIFO lanes location Size of FIFO Lane Befll10 Comments ayinetr Before Hot 3 x 7 x (100 60)% 3.6 = ( straightening Before OD grinding - pe3) (4 pieces) .5 3 x7x6 1.5 = 3.6 (4+ 5 = 9 pieces) 3efor7 xOD60% (2.1+ 2) (3 x 7) - (2 x 7 x 25%) 1.5orx = 3.5 30% (4 pieces) BeforeHettreatment 2x 7 11.6 (1 batch = 16 pieces) Before Hardness test 1x 7 Before Cold straightening pbecause ce 3 x 7 x 70% - (3 - 1) x 7 0.5 13 Before OD drift (13 pieces) Before 1.2 Before3 Before Pressure test = 8.75 pieces) ____(9 x 7 x 50% 0.9 Scenario 1 - Maximum inventory will be reached when OD Grinding is running while MPI is idling. In addition, only 30% of the products are required to go through MIPI. Heat treatment is a batch process (batch size = 16). The batch is decided by the type of material and size the process parameters (recipe) are different. To maximize batch sizes of every cycle, inventory should queue before this process. Scenario 1 - The products from heat treatment are coming in batch so the FIFO lane should have enough space to store them. Scenario 2 - Maximum inventory will be reached when hardness test is running while cold straightening is =1 3 x 7 x 50% Full welding is a bottleneck process so it must not be starved or stop working. One day inventory is chosen because this process has a relatively low first pass yield (85%). Hence, the products that need to be reworked will also enter the FIFO lane. In addition, round products from the fit up (round) will also queue before this process. Scenario 1 - The FIFO lane should be able to store products that are being produced by full welding while there is no operator at hot straightening process. When both processes are running, the queue will decrease because hot straightening is much faster than full welding. Scenario 2 - When both hot straightening and OD grinding are running, the FIFO lane will be filled because the cycle time of the hot straightening is smaller than OD grinding. Additional 5 products will come from Radiography (45% inspection). = 11.67 (12 pieces) idling. Scenario 1 - Maximum inventory will be reached when cold straightening is running while OD drift is idling. Scenario 1 - Maximum inventory will be reached when OD drift is running while pressure test is idling. 5.4.2 Summary of Future State VSM Figure 15 shows the takt time and cycle time of the processes in the future state. Notice that the bottleneck processes have been eliminated, although the cycle times for several processes seem to be very close to the takt time. It isacceptable because the operators are shared between several processes, and so they are allowed to help other processes if needed. This operator movement will affect the operator availability of each process, hence it will affect the takt time as well. Takt Time vs. Cycle Time (Future State) E Takt Time U Oval U Round-standard U Round-guardrail U Round-integral 20.0 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 -zlkk 4, Ale Figure 15: Takt time and cycle time comparison for both oval and round products - Future state The manufacturing lead time, total processing time and waiting time of the future state are summarized in Table 14. Table 14: Summary of future state VSM of Oval product family Manufacturing Lead Time (days) Total Processing Time (days) Waiting Time (days) Current State 26.8 13.4 13.4 Future State 19.8 10.4 9.4 .......... ......... .. ...... ..... .. ........ .... ...... ---- -. .... ...... .. . ...... .... .. .... .. . .. .. .. ...... ....... ..... .1..... ....... .... II . ..... ....... ... ............ .. .1 .1 1 ... . ......... As shown in Table 14, the manufacturing lead time could be potentially cut by at least 27%. Note that the waiting time calculated here is a conservative estimation because it was based on the assumption that the FIFO lanes are full. Capacity Analysis 5.5 The capacity analysis was done to identify processes that need expansion in the future (2011 to 2013) based on the forecasted customer demand for 2011. The expansion is defined as buying additional machine or equipment for the process. Using the formula mentioned in Section 4.4.3, the maximum weekly outputs after improvement were calculated as shown in Table 15 below. In the calculation, several assumptions were made: * Available time per week = 5 x 3 x 7 = 105 hours (assuming 5 working days and 3 shift per day) * The heat treatment batch size = 16 pieces * There are 5 welding booth and 3 hot straightening benches available in the shop floor Table 15: Maximum weekly output of the assembly line (after improvement) Machine or Equipment Fit up (oval) bench Plasma cutting machine Pocket fitting bench ID welding bench Fit up (round) bench Welding booth Hot Straightening bench MPI OD Grinding machine H/T furnace Hardness test Cold Straightening machine OD Drift Pressure Test Kent Milling Machine Weighted Average Cycle Time (hour/piece) 1 0.8 0.9 2.6 1.8 1.4 0.7 0.8 1.5 0.8 0.5 1.2 0.9 1.0 8.1 First Pass Yield (FPY) % 100% 100% 100% 100% 100% 90% 100% 97% 100% 92% 92% 100% 100% 96% 100% Machine Uptime % 100% 99% 100% 95% 100% 99% 100% 100% 97% 95% 100% 100% 100% 100% 100% Max. Weekly Output (pieces) 105 129 116 38 59 65 156 135 67 116 192 84 113 101 13 ...... ........................................... .... .... ..... ............ ... The maximum weekly output then compared to the forecast weekly demand for 2011, 2012 and 2013 as shown in Table 16 below. Table 16: Maximum weekly output vs. forecast weekly demand for 2011, 2012 and 2013 (in pieces) Fit up (oval) bench Plasma cutting machine Pocket fitting bench ID welding bench Fit up (round) bench Welding booth Hot Straightening bench 105 129 116 38 59 65 156 102 2012 Weekly Demand 103 103 103 19 19 136 136 MPI 135 29 38 Machine or Equipment Max. Output 2011 Weekly Demand 77 77 77 14 14 2013 Weekly Demand 116 116 116 23 23 155 155 43 67 91 H/T furnace 116 99 133 Hardness test 192 99 133 Cold Straightening machine 84 91 OD Drift 113 91 122 139 Pressure Test 101 95 128 145 13 9 12 15 OD Grinding machine Kent Milling Machine 139 152 139 If the difference between the weekly demand and the maximum weekly output is less than 50% of the maximum weekly output, Kaizen or improvement should be done to the process to increase its output. However, if the difference is larger than 50%, expansion is most likely required. However, the result of the improvements will also depend on the state of the process. If the process is already in an optimized state, it will be difficult to improve the process any further. As shown in Table 16, the yellow color indicates that the process needs to be improved to meet the forecast demand. On the other hand, the red color indicates that the process might require expansion; this will depend on the result of the improvement done previously (if any). For example, the fit up (oval), OD drift and Kent Milling machine do not require expansion because the difference is less than 30%; however Kaizen or improvements need to be done to increase the output of these processes. The summary of the capacity analysis for the rest of the processes is shown in Section 6.2. Chapter 6: Recommendation 6.1 Recommendation on the Current State The following section describes the suggestions to improve the current state of the manufacturing line. 6.1.1 FIFO Lanes Implementation The FIFO lanes with maximum sizes should be implemented between the processes as shown in Figures 13 and 14. If the FIFO lane is full, the upstream process of that lane has to stop and the downstream process of the lane has to start and clear the queue. Products that need to be reworked are allowed to cut the queue and they are counted as inventory inside the FIFO lane. Implementation of the FIFO lanes has the following advantages: * Limit WIP inside the system * Give signal to operators when to stop and move to another station or process * Ensure that the work order that was released earlier is being prioritized and finished first The maximum size of the FIFO lanes can be reduced in the future if all the processes are continuously running at the same time or dedicated operators are assigned to the processes. 6.1.2 Production Scheduling First, the work orders should be released daily according to the takt time of the bottleneck processes (full welding). Daily release of work will minimize WIP to one day inventory in the upstream process (before fit up, plasma cutting and pocket fitting). The feasibility of implementing this one day inventory will need to be further discussed with the warehouse department. Second, the load should be leveled and sorted according to the heat treatment recipes. It will help to create more efficient scheduling and also minimize the number of multiple FIFO lanes that are needed before the heat treatment process. 6.1.3 Fit Up, Plasma Cutting and Pocket Fitting Assembly Cell In the future state the assembly cell should be implemented for the fit up, plasma cutting and pocket fitting. Re-layout of the booth is needed to allow efficient movement of operators and product being processed. The advantages of using the assembly cell concepts are: * Create a continuous one piece flow between these three processes e Eliminate WIP between fit up, plasma cutting and pocket fitting " Reduce processing time by minimizing loading and unloading of product to the work station or bench Currently, the fit up, plasma cutting and pocket fitting are done inside the oval fit up booth. In every shift, there are one or two operators working on the booth. Commonly, the fit up is done in the 1 s' shift while plasma cutting and pocket fitting are done in the next shift. As shown in Figure 12, the fit up (oval) is a bottleneck process, and so, the outputs of these three processes are limited by the fit up (oval) process. 6.1.4 Heat Treatment Improvement Several improvements that should be implemented at the heat treatment process are: * Total Productive Maintenance (TPM) The current uptime of heat treatment furnace is only 85% and most of the breakdown takes very long to repair. Total Productive Maintenance (TPM) should be implemented in order to increase the uptime of the furnace. Currently, the heat treatment station only works 2 shifts so the 3 rd shift could be used to do the maintenance work. * Multiple FIFO lanes and heat treatment schedule The products should be sorted based on the recipe of the heat treatment process. In addition, scheduling of the heat treatment process should be implemented to further increase efficiency. 6.1.5 Workload Balancing MPI, hardness test and pressure test are done by the inspectors. Instead of having a dedicated inspector for each of the inspection processes, workload balancing is proposed. One solution will be to shift the working hours of the MPI operator so that he could work during the 1st and 2 nd shift. For example, in the 1' shift he will work on the MPI station and when the 2 shift starts, he will work on the pressure test. 6.1.6 Other Improvements First, the number of products being shipped out to suppliers should be limited to 10 pieces. This is recommended to ensure that the suppliers for threading, sandblasting and painting could commit to the 7-working-day lead times as shown in Table 21 in Appendix D. As an illustration, if 11 pieces are being shipped out to the suppliers, the lead times might extend to 13 working days. In addition, it will ensure daily delivery to the suppliers, because the daily output according to the takt is about 9 pieces per day. For the future state VSM shown in Figure 14, the supplier lead time was assumed to be 9 days, taking into account additional 2 days for logistics. In the current state, as shown in Table 11, the average number being shipped out to the suppliers and the average lead time do not match the quoted lead time given by suppliers. There is an opportunity to cut the supplier lead time. In addition, the company might want to consider bringing the processes in-house, if possible. Second, basic lean tools such as 5S should be implemented in the assembly line to increase the efficiency of the worker. It will reduce the cycle time by minimizing non-value added activities, such as looking for tools, work order, etc. In addition, visual management such as production control board should be updated every shift by every worker to keep track of the work orders progress and increase communication across the line. 6.2 Recommendation on Capacity Expansion In this section, actions that need to be taken are listed and suggested based on the analysis in Section 5.5. The processes that require expansion are: " Full Welding The current capacity of full welding process is 5 welding booths. In order to meet the 2011 customer demand, a total of 8 welding booths are needed. In order to meet the 2013 customer demand, a total of 12 welding booths are needed. The company might consider doing one time expansion to prepare for next three years. " OD Grinding In order to meet the 2011 forecast demand, Kaizen needs to be done to reduce the cycle time by 30%. However, the addition of one OD grinding machine will be required in 2012. * Heat Treatment The addition of one heat treatment furnace will be required in 2013. Kaizen to increase the machine uptime also needs to be done to meet the 2012 forecast demand. However, because the heat treatment furnace is unstable and takes a long time to repair, the company might want to have extra capacity for these processes as backup. It is recommended to have the second furnace running in 2012. * Cold Straightening In order to meet the 2011 forecast demand, Kaizen needs to be done to reduce the cycle time by 15%. In order to meet the 2012 forecast demand, 33% cycle time reduction is needed. However if it is not achieved, one additional cold straightening bench will be required in 2012. * Pressure Test In order to meet the 2012 and 2013 forecast demand, Kaizen needs to be done to reduce the processing time by 20% and 30% respectively. If it is not achieved, then additional pressure test bay might be needed in 2013. Chapter 7: Conclusion and Future Work 7.1 Conclusion In this project, material and information flow of the manufacturing line have been captured using Value Stream Mapping. The current manufacturing lead times have been identified which are 27 days for oval products and 56 days in average for round products. Fit up (Oval), full welding and pressure test were identified as bottleneck processes using takt time analysis. Recommendations have been proposed to eliminate these bottlenecks and improve the performance of the manufacturing line. The improvements proposed were to implement FIFO lanes, assembly cell, workload balancing, and Kaizen activities. The future VSM shows that there is an opportunity to cut the manufacturing lead time by at least 27% if these recommendations were to be implemented. In addition, to ensure that the manufacturing line will be able to meet the forecasted demand for 2011, 2012 and 2013 capacity analysis was done to identify potential bottlenecks in the process. These will help the company to decide when to initiate Kaizen activities and when to buy new machines or equipment for a particular process. 7.2 Future Work After the recommendations have been implemented, value stream mapping needs to be done to measure the new 'current state' of the manufacturing line in the future. New waste and bottlenecks might be identified in the future, and they have to be eliminated. These efforts must be done as part of continuous improvement activities in order to pursue perfection of the manufacturing line. Secondly, simulation could be done in order to determine the optimal size of the FIFO lanes between processes. The maximum size of the FIFO lanes will be reduced further when dedicated operators were assigned in every station ensuring more continuous flow. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] What is Lean. Lean Enterprise Institute, Inc.; 2009. Available from: http://www.lean.org/whatslean Womack, J. P., Jones, D.T., and Roos, D. (1990). The Machine That Changed the World: The Story of Lean Production. Harper Perennial; 1991. Womack, J.P. and Jones, D.T. Lean Thinking. 2nd ed. New York, NY: Free Press, Simon & Schuster, Inc.; 2003. Recker, R. and Bolstorff, P. Integration of SCOR with Lean & Six Sigma. Advanced Integrated Technologies Group, Inc.; 2003. Available from: http://www.scelimited.com/sitebuildercontent/ sitebuilderfiles/ scorsixsigmaconvergence.pdf Lean Production I Business Studies Theory I Business & Marketing Resources. The Times from: 2010. Available Ltd.; Publishing MBA Ltd. and Newspapers http://www.thetimeslO0.co.uk/theory/theory--company--252.php Thompson, S. W. Lean, TOC or Six Sigma. Society of Manufacturing Engineers; 2010. Available from: http://www.sme.org/cgi-bin/get-newsletter.pl?LEAN&20030811&2 Rother, M. and Shook, J. (2003), Learning to See: Value-stream mapping to create Value and eliminate Muda by, The Lean Enterprise Institute. Duggan, Kevin J. Creating Mixed Model Value Streams: practical Lean techniques for building to demand. New York: Productivity Press, 2002 Dennis, P. Lean Production Simplified: A Plain Language Guide to the World's Most Powerful Production System. New York: Productivity Press, 2002. Bo, Li (2009), Making a High-Mix Make-to-Order Production System Lean, Master Thesis, Massachusetts Institute of Technology, Cambridge, MA. Marek, Richard P., Elkins, Debra A., and Smith, Donald R., Understanding the fundamentals of Kanban and CONWIP pull systems using simulation, Proceedings of the 2001 Winter Simulation Conference Haiqing, Guo (2009), Head & Base Production Optimization: Setup Time Reduction, Master Thesis, Massachusetts Institute of Technology, Cambridge, MA. A study of the Toyota Production System, Shigeo Shingo, Productivity Press, 1989, p 47 Software available from: http://www.igrafx.com . ............ ....... .... APPENDIX A - VSM Icons MATERIAL FLOW ICONS C/T = 45 sec. CO= 30 min ASSEMLY 3 shifte ;XYZ Manufacturing Process Mon. +Wed. Truck Shipment 300 pieces I Day Outside Sources Data Box may 20 Finished Goods to Customer PUSH Arrow Buffer or Safety Stock loxoxi Schedule 4. Electronic Information Flow 7 Signal Kanban Kanban Withdrawal Operator INFORMATION FLOW ICONS Withdrawal First-In-First-Out Sequence Flow 0 Supermarket Manual Information Flow piw. -FIFO- wxxxxxb GENERAL ICONS Kaizen Lightning Burst Inventory Load Leveling Y Kanban Post Production Kanban Sequenced-Pull Ball Kanban Arriving in Batches Figure 16: VSM icons [71 "Go See" Scheduling APPENDIX B - Product Family Selections Group A Group B Group C Group D Group E Figure 17: Previous Work of Product Family Selections Oval XX Round-standard Round-guard rail Round-integral Figure 18: Current Product Family Selections 60 APPENDIX C- Current State VSM for Round Product Families 0.8 25.0 days | days 0.2 days 1.4 days | 0.1days | | 0.0 days | | 0.4 days | O.4 days 1.0 days | 0.1 days | | 4.0 days Figure 19: Current state VSM of Round-standard product family (1/2) 0.9 d 1.7 days 0.1 days || 0.6 days Customer Demand: 3 pieces per Week (rakt Time 35 hours) 0.9 days ,s 0.8 days 0.0 days 0.9 days 0.1 days 0.8 days 0.1days 3.1 days 0.0 days 1.0 days 0.0 days 12.0 days Figure 20: Current state VSM of Round-standard product family (2/2) Manufacturing LT = 55.5 days Total P/T = 42.5 Waiting Time = 13 days 2days 0.84days -- - _25.0 days | 0.1days |- I| 01 days | 0.4days | 1. days | 0.1 days|p .4days 4,0dayp 1.7days o.1days | 0.9days |. Figure 21: Current state VSM of Round-guardrail product family (1/2) 63 0.8days 08 days .9days | | days | 1days DA | | Oday Cuwner Demand: 3 piecesperWeek (TaidTime36 hors) days o.0 dys 0.9 das | | .3days _j------L 0.3 days | -- L | 02 days | 0.0days-| 3.1 das 09 days 0.0 days | --L| 0.0 days | 0.0days 10 days | | Figure 22: Current state VSM of Round-guardrail product family (2/2) ManufacurugLa =7.6 days 12.0 days 'Tota Procssin Tonw= 43,7 days lWaitingTimei= 13.9 days 0.8 days 1.4 days 0.2 0Odays days . 0.4 days Figure 23: Current state VSM of Round-integral product family (1/2) 65 1.7 days s09 days 0.8 CustomerDemand: 3 pieces per Week (TaidTime35 hours) das8d 0.8 0.1 d0. 0.1 days 0.0 days 1.0 days days d3.1 10,6 0.1days 05 days |0.0 days | 0.0 days | 120 days Figure 24: Current state VSM of Round-integral product family (2/2) Mansufactring L/T= 56.3 days Total Processing Time= 433 days Waiting Time= 13 days APPENDIX D - VSM Data Table 17: Processing time of oval products (in hours) Process 27/8 Fit up (Oval) Plasma Cutting Full Welding Hot Straightening HHT 4130 1.1 0.9 4 1/2 1.5 1.7 27/8 0.7 0.9 1.5 1.4 Heat Treatment Cold Straightening _ _ _1_ 1.3 4 1/2 1.6 27/8 7.5 4.7 31/2 8.1 6.5 41/2 9.4 6.5 27/8 1.5 2 3 1/2 1.9 2.1 2.7 1 8.3 2.8 1.5 1.5 3 1/2 41/2 27/8 1.7 17.5 1.3 9 2 10 3 1/2 17.5 9 10 41/2 17.5 9 10 27/8 0.8 1.5 1.1 3 1/2 41/2 1.1 1.4 27/8 0.8 1.2 41/2 1.5 0.6 3 1/2 OD Drift _ 1.3 1.4 1.3 1.3 3 1/2 27/8 1.2 1.1 3 1/2 41/2 OD Grinding LHT 4130 1.3 3 1/2 41/2 27/8 Pocket Fitting 410SS 0.9 1.2 1.2 . .. . . ...................... I........................ . ....... .. -- - Table 18: Processing time of round products (in hours) Process ID Welding Fit up (Round) Full Welding Hot Straightening OD Grinding Heat Treatment Cold Straightening ODDrift 410-13Cr Super 13Cr Inconel 925 2.8 1.6 2.9 41/2 51/2 2.5 4 1/2______ 51/2 41/2 7.3 41/2 1.6 41/2 41/2 1.15 5 1/2 5 1/2 51/2 7.5 12.5 _____ 12.5 1.6 1.5 1.4 1.7 12 32 41/2 51/2 17.5 17.5 12 12 32 32 41/2 1.4 51/2 1.5 41/2 1.2 51/2 1.1 1.7 0.8 Table 19: Number of products queuing before processes (oval products) 5/24/2010 5/25/2010 5/26/2010 5/27/2010 5/31/2010 6/3/2010 Average Fit Up (Oval) 8 36 13 26 12 16 20 Plasma 0 7 0 0 0 4 2 Pocket Fitting 0 1 0 0 0 1 1 Full Welding 11 9 13 12 4 11 10 Hot Straightening 0 4 4 5 10 5 5 OD Grinding 0 5 5 2 6 7 5 MPI 8 4 6 10 4 7 7 Heat 13 20 4 9 36 0 14 14 10 12 20 5 16 13 Cold Straightening 5 0 2 0 2 14 4 OD Drift 0 5 12 0 5 4 5 Pressure Test 23 29 34 27 2 8 21 Total 77 131 105 111 86 88 107 Date Cutting Treatment Hardness Test --- - ------ Table 20: Number of products queuing before processes (round products) Date IDWelding Fit Up (Round) Full Welding Hot Straightening OD Grinding MPI Heat Treatment Hardness Test Cold Straightening OD Drift Pressure Test Total 5/31/2010 3 0 0 0 0 0 1 0 0 1 0 4 6/1/2010 0 0 0 0 0 0 1 0 1 3 2 7 6/2/2010 0 0 1 0 0 0 3 0 0 4 2 10 6/3/2010 6/4/2010 8 0 0 0 1 0 0 0 0 0 0 0 8 3 0 0 0 0 0 3 0 1 17 7 Table 21: Lead time of outsourced processes quoted by suppliers Outsourced Process Threading Quantity 5pcs Lead Time 3 working days 5pcs - 10pcs 3 working days 11pcs - 20pcs > 20pcs 5 working days 8 working days 5pcs 5 working days 5pcs - 10pcs 5 working days 11pcs - 20pcs 10 working days > 20pcs 15 working days 5pcs 2 working days 2 working days 5pcs - 10pcs Sandblasting and Painting 11pcs - 20pcs 3 working days > 20pcs 5 working days Note: Working days excluded Sat, Sun and Public holiday Average 2 0 1 0 0 0 4 0 1 3 1 13