This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Multinational companies need to achieve high performance in productivity and quality regardless if the plant is located on developed or on emerging countries. And sometimes, the kind of obstacles and opportunities are differents in each site, mainly when the high technology in developed countries is not completed transferred to subsidiaries in emerging countries.
This paper researched 30 manufacturing cells from an automotive multinational company established in Brazil, identifying 16 characteristics used in operation of equipments under a lean production system and observing its efficiency. Additionally was applyed the software (SPAD) to nonparametric Multivariate Analysis of Data that showed the influences that each of the 16 features have on the manufacturing cells and grouped into classes.
The conclusion was that although the technology used is not the most current in global market, there is scope for improving productivity by working on some characteristics of the lean model.
Key words : Overall Equipment Effectiveness OEE, cellular manufacturing, productivity, manufacturing performance
Along the last three decades, companies have radically changed the structure of manufacturing as a response to the globalization of markets and increased competitiveness. The strategy is to establish local plants with similar production structures to the model of excellence in global operations. Reference systems are being adopted by global companies in order to form and establish a common language that enables the exchange of production flows between plants. Systems called World Class Manufacturing and Lean Production Systems are being adopted to achieve this goal.
An element used in this effort to achieve operational excellence is the measurement of efficiency losses of productive assets through operational indicators. The methodology for measuring losses known as Overall Equipment Effectiveness (OEE) has become consecrated, an indicator used to measure the losses within their productive operations on global companies. The use of the indicator OEE allows companies to analyze the actual efficiency on the use of their assets. The analysis of these conditions starts from the identification of losses existing in manufacturing environment. Then, it measures the level of equipment availability, performance and quality and finally it generates an overall indicator that measures the efficiency of the asset utilization through the relationship between the effective time of adding value activities and the total available time in a given period (LJUNGBERG, 1990; NAKAJIMA, 1988).
However the measurement of the losses of the equipment has not yet been described in the literature about the characteristics of use of equipment in a particular production system.
Through the use of a standardized measurement calculation of losses known as OEE, the research was designed to achieve the following objectives:
Evaluate the degree of influence of the operating characteristics of the production system on the use of the equipment adopted by company in a lean management environment and the respective results of efficiency of the equipment.
Compare the efficiency between the equipment taking into consideration the clusters of characteristics identified.
Identify the characteristics of use of equipment in production system which can yet to be improved by adoption of lean tools.
2 THEORETICAL BACKGROUND
2.1 Models of Production Systems in pursuit of Excellence
The system of mass production in Fordism has grown with the principle of producing large quantities of products or parts to meet the growing demand for standardized products in the first three quarters of the twentieth century. In the period when mass production was sovereign, the decision to produce in large lots was not a problem because inventory costs and long lead times were absorbed by the demand for large quantities of a few models. When demand began to break down into many different types of small quantities each, a new scenario was being created where the simple production of large lots was no longer the best solution (WOMACK, JONES, ROOS, 1990).
In the system of mass production, the cost decreases as the production volumes increases (economies of scale). In postwar Japan, particularly at Toyota, the need to replace the model of mass production by another model, more suitable to the needs of that company to produce small batches of a high variety of vehicles with different characteristics (WOMACK, JONES, 1996) brought the lean production model to life. In the system of lean production, the cost per unit is not reduced by the growing volume of production, but by eliminating the waste or loss along the production flow.
This new form of Manufacturing Management became popular with the publication of the book "The Machine that Changed the World" (WOMACK, JONES, ROOS, 1990). The waste elements are classified into seven types: overproduction, waiting, transportation, processing, inventory, movement and production of defective items (OHNO, 1997; WOMACK, JONES, ROOS, 1990 SLACK, 2002) Reducing these losses and waste production means looking for operational excellence characterized by high performance efficiency and productivity. In the model of lean manufacturing the goal to achieve excellence is supported by two pillars.
The first pillar is JIT (Just-in-time) with the principle of producing what is needed only when needed. This pillar introduces the concept of pull production as a way to avoid losses linked to large batch production above demand. This loss is not accounted for directly by the OEE, however is the increased availability and efficiency of productive assets that enable the production of JIT because they increase the speed of response to the amount of demand.
The second pillar is autonomation (Jidoka) and means an automation with a human touch or automation humanized (OHNO, 1997). One way to practice autonomation is to separate man from machine to allow it to act on multiple machines, always having the authority to stop the line in case of occurrence of defects. Another way is through the deployment of devices to test errors and failures, also known as poka-yoke or baka-yoke (OHNO, 1997; SLACK, 2002). In lean production, the physical arrangement or layout recognized as a facilitator of these two pillars and therefore the identification of losses is the cellular arrangement. A manufacturing cell is usually shaped in a "U", because experience has shown that this shape achieves the best results in terms of productivity, and facilitates both the communication between traders and the visualization of the whole cell. In this physical arrangement, the number of operators is 50% to 70% lower than the number of existing machines (GAURY, KLEIJNEN, PIERREVAL, 2008). With the approach of the machines, the distances and routes of operators decrease proportionally, allowing the operation of several machines by one operator.
The time cross of the cell is on average 90% lower than the functional physical arrangement (HINES, TAYLOR, 2000)
Waste in production systems
Waste can be defined as any operation or activity that takes time and adds no value to the process. Toyota has identified seven types of waste (OHNO, 1997) on their production system, possible to be applied to the industrial reality on other organizations:
Overproduction - Production greater than the amount demanded by the next levels, is the biggest source of waste;
Wait Time - The amount of time spent on waiting for materials which occurs when operators are producing parts in the process not at the necessary time;
Transportation - The movement of materials within the factory as well as excessive movement of the stock in process, is an activity that adds no value;
Process - The process itself, there may be sources of waste. Some operations only when there are deviations and failures of projects in the maintenance, so they are likely to be eliminated;
Inventory - Inside the lean philosophy, all stock becomes the target of elimination. But only one can reduce inventory by eliminating its causes. You must remove all the stock bumper. Besides plastering a considerable amount of capital stocks delay feedback on the quality, which means more waste and more rework;
Handler - Sometimes, the operator may seem busy, but is only moving because of flaws in the process. Derive there from the costs related to the drive useless in achieving the activities, ie, due to the inefficiency of the operation itself;
Defective Products - are products out of specification. The noise created by defective parts along the process can upset the entire system. Some devices that reduce the incidence of these failures are called poka-yokes.
Ranking the seven classic types of losses and defining them as an unnecessary work (actions that generate costs, but do not add value to the product / service), Ohno (1997) declares that before recognize them it is necessary to understand its nature.
It also defines lean manufacturing as a systematic approach to the elimination of waste in a process of continuous improvement in pursuit of perfection from the needs of customers.
Hines & Taylor (2000) suggest that we need to equip workers with "glasses changes" (changes in the Japanese language means any activity that consumes resources without adding value to customers), enabling them to see losses. The idea is to create a culture that encourages them to eliminate the losses identified. Achieving this goal requires not only the qualitative understanding of the categories of waste, but also a measurement system that allows to assess to what degree these wastes occur in the system reaching certain levels of losses.
Shah and Ward (2003) related to lean production the relevant practices, through the vision of many authors. Among these practices, one can find the production level, cell manufacturing, Kanban, cycle time reduction, batch size reduction, competitive benchmarking, Quality Management programs, among others. Each of these authors cited by these researchers made â€‹â€‹reference to some of these practices.
Performance indicators of production systems
Key Performance Indicator - KPI metric is the quantitative indicator of efficiency, performance or achievement. Are vectors of measurable performance currently used by companies to measure results and decision - making stand upon short, medium and long term.
One of the key factors to business excellence is to have production systems and efficient operations, enabling organizations to gain competitive advantages over their competitors. In this context, the success or failure of many companies can be explained by the harmony between competitive strategies, manufacturing technologies and ways of Production Management (SLACK, 2002).
One way to know if the production system is in line with company strategy and acting as a booster agent of competitive advantages over its competitors is through an adequate system of performance evaluation for the production function. The performance measurement system for the production function for the range of indicators and reports that aim to assess how production is going in relation to the targets of the company's strategy.
Johnson and Kaplan (2001) show that the use of indicators of non-financial performance is to evaluate the best monthly performance of a company. They claim that the use of financial indicators no longer reflect their recent performance. They claim that can be challenged by rapid changes in technology, reduced life cycles for products, by innovations in the organization of production operations and the inclusion of expenses from previous periods or those that include benefits that are achieved in the future. Allow non-financial indicators and set targets to better predict the long-term profitability of the company.
Johnson and Kaplan (2001) continue that this picture is justified by the need for further evaluation of the performance attributes of companies that effectively reflects the integration and flexibility of their resources. It can be concluded that the performance is manageable in the proportion that is measured. Without these measurements, managers can not communicate specifically what the performance expectations and what the expected results of subordinates.
Introduced by the asset management system known as TPM model (Total Productive Maintenance ) OEE is a set of indicators used to measure performance of the productive system in relation to their resources of production, enabling companies to analyze the actual conditions of use of their assets (DAL,TUGWELL,GREATBANKS,2000).These conditions are analysed from the identification of losses existing in the manufacturing environment, involving rates of availability of equipment, performance and quality.
Measuring the overall effectiveness of the equipment can be applied in different ways and goals. According to Jonsson and Lesshammar (2005) the O.E.E. indicates areas where improvements should be developed or can be used as a benchmark indicator, allowing developed to quantify the improvements in equipment, production lines or cells over time.
According to Nakajima (1988) the O.E.E. is a measure that seeks to reveal the hidden costs in the company. Ljungberg (1990) show us that before the advent of this indicator, the availability was only considered in the use of equipment, which resulted in overestimation of capacity. According to Nakajima (1988) the O.E.E. is measured from the stratification of the six big losses and calculated by multiplying the indices Availability, Performance and Quality. Also according to Nakajima (1988) O.E.E. of 85% should be sought as the ideal goal for the equipment. Companies that had an O.E.E. above 85% won the prize TPM Award. To obtain the value of OEE, it is necessary to calculate: (90% for availability X 95% for performance X 99% for quality).
Figure 1 illustrates the indicator OEE - its contents and losses related to each one.
Figure 1 - Indicator O.E.E.
Source: Adapted from Nakajima 
The O.E.E. index can be calculated through the expressions of the Availability, Quality and Performance (speed) Indexes, as described below (JONSSON, LESSHAMMAR, 2005)
Availability Index: Answers the question of whether the machine is running. For that, they consider the following losses:
Losses due to management issues (waiting for programming, operator failure, lack of tooling, shortage of blanks, etc.)
Losses due to unscheduled downtime (maintenance, setup, pending award, power outages, etc.)
Equations (1), (2) and (3) refer to the calculation of the availability index:
Performance Index (Speed): The second index is a response to the following question: "Is the machine running at the planned speed?" This index can be obtained from equation (4).
To Nakajima (1988) the difference between theoretical and actual performance occurs due to losses related to downtime and drop of performance (drop in the speed to which the machine was designed).
Quality Index: The third index that composes the OEE answers the question: "Is the machine producing with the right specifications?" This index can be obtained from equation (5).
Overall Equipment Effectiveness (OEE): OEE indicator, as previously mentioned is composed of three previous indexes. According to Jonsson and Lesshammar (2005) its objective is to analyze the effectiveness of the equipment and not the operators. Therefore, it is used to verify that the machine continues to work on speed and quality specified in your project and also to point out the losses from the production system as a whole. This index can be obtained from equation (6).
O.E.E. (%) = Availability x Performance x Quality (6)
Therefore, the identification of losses (waste) is the most important activity in the process of calculation of OEE the limitation of the company to identify their losses in the act prevents the restoration of original equipment, ensuring effective global reach, as established when the product was purchased or renovated. The O.E.E. is quoted by Kenyon, Canel and Neureuther (2005) as one of the key factors to reducing costs, increasing productivity and hence the net. The authors evaluate the impact of batch sizes on factors that directly influence the OEE, as quantities produced, in-process inventory, profit, operating costs and lead time.
There is no consensus on what is the ideal minimum OEE index and the literature values â€‹â€‹are from 30% to 85%. The most currently accepted value is 85%, which was established by Nakajima (1988). It is composed of 90% availability, 95% efficiency and 99% quality (0.90 X 0.95 X 0.99 = 0.85).
The research was divided into three stages with triangulation between them. In the first stage was developed with the emphasis on quantitative measurement of OEE indicators cell machinery in the second stage the emphasis was through qualitative analysis of the characteristics of equipment for use in lean production system, and finally the third stage, turned to trying to identify the quantitative characteristics of these influences on the overall efficiency.
3.1 The company used as object of study
The company elected to carry out this research is an auto company in the city of Sâo Bernardo do Campo in metropolitan Sao Paulo, Brazil's most industrialized region. The company is a global organization present in several countries in Europe, North America, Latin America, Africa, Asia and Oceania, employing 273,216 employees worldwide (2010 year). This company is based in Stuttgart, Germany.
In Brazil, it has three units that have a total of 14,073 employees (2010 year). The unit located in São Bernardo do Campo employs 11,986 workers (2010 year) and has a large industrial facility where you can find manufacturing activities such as machining and assembly of mechanical components, activities related to the development of new products and technical support areas for across the enterprise, offering computer services and maintenance of its infrastructure.
From 1994 to 2000 the Brazilian unit has introduced a set of practices compatible with the lean production model, effective practices of Kaizen, 5S, TPM, manufacturing cells. In the years from 2004 until 2009 the corporation began the global standardization of these practices and began to adopt the OEE as a tool that enabled the comparison of results from process efficiency and the guided management of resources. In 2004 and 2006, the Brazilian unit received the award of the prize for best plant in the application of lean concepts, making it influential in global perspective of the corporation.
3.2 Characteristics of the analyzed sample
The 30 manufacturing cells considered in this study belong to the production sector of the front and rear axles of commercial vehicle of the brand, with 893 machines (2010 baseline) distributed in a pavilion of 71,650 square meters with approximately 1,500 employees. Table 1 displays some indicators of the sector in which the cells are located.
The shop floor is composed by the latest technology equipment and also with equipment with more than 35 years (figure 2). In all devices, predictive, preventive and corrective maintenance concepts are applied.
Total Area (m²)
Built-up Area (m²)
Annual Production (Axes)
Productivity / Employee
Primary expenses (Mio)
Note Global Standardization
Table 1 - Annual Indicators of Manufacturing Board of Axes
Source: "adapted from" Report of the company studied.
The list of equipment includes items that perform the machining, heat treatment and assembly of various parts that compose the axle. The machines are aligned according to the concept of manufacturing cell and work with the schedule set by the system "kanban.".
The sample of 30 manufacturing cells was set up 15 years ago, and is operated by structured work teams (Team work).
The production system of these cells in the study consists of machine tools arranged in the cell type and layout adopted in the form of JIT logistics.
The approval of the quality of work is the responsibility of the operator, which characterizes the autonomous work teams (Team Work).
The cell efficiency is measured by calculating the OEE of the bottleneck of the cell, one that limits the productive capacity of the group of machines.
The O.E.E. is calculated in the company studied in particular, using the indicator on the overall result of equation (8) and categorizing the losses in an specific relationship, described in a collection form and recorded in a database.
Figure 2 - Age of Machinery Manufacturing Board of Axes
Source: "adapted from" Report of the company studied.
The data of daily efficiency of the bottlenecks from the 30 cells shown in Table 2 was collected during a period of 17 months in the years of 2009 and 2011.
Gross Available Time
Net Available Time
Average Produced Time
O.E.E. (average) Net
Complete axle housing assembly
Wheel Hub HL5
Complete axle housing assembly
Wheel drum HL7
Steering knuckle VO4
Heavy duty brake shoes
Wheel drum HL7
Heavy duty wheel disk
Steering knuckle LN
Steering knuckle VL3
I-Beam VL3 6,5 T
Wheel Hub VL2/HL2
Wheel Hub HL7
Differential housing HL5/HL4R
Wheel Hub HL4
Wheel Hub VL3 5T
Medium duty axle housing
Medium duty break shoes
Wheel hub VO4/VL4
Wheel hub M2000
Wheel drum VL3
Medium duty break shoes
Table 2 - Manufacturing Cells with O.E.E. Average for the Period
The cells were grouped into three categories according to their average efficiency:
10 cells with low performance, is below 75%,
Middle performance 11 cells, is between 75% and 85%,
High performance 09 cells, were, above 85%,
3.3 Characteristics of equipment for use in cells in the lean production system.
The second stage of the research started with over production operating systems the technical literature about characteristics directly or indirectly linked to the equipment. 15 features were identified which generated 32 possible applications:
Feeding process is automatic or manual;
Team work is available or not;
Celular layout is available or not;
The bottleneck is a CNC or conventional machine;
Operator dedicated to one single machine in the bottleneck, or he operates more machines;
The work station has ergonomic issues or not;
Preventative maintenance is in place or not;
Statistical process control (SPC) is available or not;
Type of machined material: steel or cast iron;
Age of the bottleneck equipment is older than 0, 5, 10, 20 or 30 years;
One or more products is(are) manufactured in the location;
Weight of products;
Weekly number of setups;
Average time spent on each setup;
Number of operators working in the cell.
Later, each cell was analyzed considering the presence - a factor - or absence - 0 factor - of these characteristics in the equipment bottleneck (table 3).
Table. 3 - Check list of Characteristics of the Cells - Source: Authors.
4. ANALYSIS OF DATA
4.1 Influence degree of use characteristics of equipment within the lean system over the efficiency of the manufacturing cells.
The study of influence on the overall efficiency began with the mapping of individual characteristics of each cell. Each feature was evaluated by their relative participation on each group, established in the first phase of the study as low, medium and high performance.
High performance Group = consisting of 09 cells with performance greater than or equal to 85%;
Middle performance Group = consisting of 11 machines with performance between 75% and 85%;
Low performance Group = consisting of 10 machines with performance lower than or equal to 75%.
At this stage the features for which percentages appeared simultaneously in two or three groups with variations of the order 0 to 20% were eliminated from further analysis. The elimination is based on qualitative correspondence between the low presence of the characteristic and the change of the degree of efficiency.
Selected 10 characteristics that most influence the performance of the manufacturing process of the sample that generated 16 most influential possibilities seen in table 4.
Loading process is automatic or manual;
The bottleneck is out of cell layout;
The bottleneck is a CNC machine;
The bottleneck is simple or complex conventional machine;
The bottleneck has ergonomic interference;
Preventive maintenance is in place;
Type of machined material - steel or cast iron;
Age of the bottleneck is higher than five years;
Average time spent on each setup;
Number of operators working in the manufacturing cell.
Features Present in the Cells Studied Groups
Low Performance (10 cells)
Middle Performance (11 cells)
Preventive maintenance applied
Type of Material = Steel
Age up to 5 Years
Set Up - Up to 10 min
Set Up - Up to 60 min
Set Up - Beyond 60 min
No Set Up
Up to 2 Operators Case
Up to 3 Operators Case
Beyond 3 Operators Case
Table 4- Summary of Attendance Percentage of 16 features per manufacturing cell
Statistical Analysis Data
The statistical technique employed in data analysis was Correspondence Analysis (GREENACRE,2003;1984), a non-parametric methodology. Correspondence analysis is a statistical method of Multivariate Analysis of Data. According to Hair et al. (2005) Multivariate Data Analysis refers to all statistical methods that simultaneously analyze multiple measurements on each individual or object under investigation.
This method is applied to large cross tables or contingency tables, which are basically used for discrete variables. It can also be applied to continuous variables, according to Hair et al (2005), provided they are properly coded. It is generally used when there is interest in studying the relationships between sets of variables that are being crossed. After doing the analysis, if the relationship between elements of different sets of variables, one can say that these sets are in correspondence. The Correspondence Analysis involves two sets of variables called cross-Binary Correspondence Analysis, as in the case of more than two sets of variables cross, called Multiple Correspondence Analysis.
As a research tool, we used the Access database OEE the company studied. Their numbers were processed with the help of programs like Excel and SPSS 11.0 (to obtain the parameters of the descriptive statistics of the data) and SPAD 3.6 - Portable Système pour l'Analyse des Donnes (for the procedure of nonparametric Multivariate Correspondence Analysis). Applying the SPAD show the influences that each of the 16 features have on the manufacturing cells and grouped into classes.
The first attempt to cluster in three classes generated a high degree of dispersion within each cluster. There was the suggestion of a new grouping with four classes which allowed a lower dispersion:
Class 1 / 4 - Composed of 04 cells;
Class 2 / 4 - Composed of 09 cells;
Class 3 / 4 - Composed of 09 cells;
Class 4 / 4 - Composed of 08 cells.
Figure 6.14 shows graphically the results of the Table of the Results of grouping the four classes of cells that held 3.6 SPAD program. Each class had the following characteristics linked to efficiency standards:
From the 04 cells that make up this group, 03 are high performance. All three have in common:
Automated workstation, load/unload is automatic and does not depend on the operator.
Preventative maintenance is practiced.
From the 09 cells that compose this group, 05 are presenting medium performance. They all have two things in common:
Workstation "multi machine", where one operator works in two or more machines.
Setup with average time of 60 minutes duration.
From the 09 cells that compose this group, 06 are low performance. All three have in common:
Four or more operators that interfere with the production, i.e., the cell is composed of several employees.
Setup medium with time over 60 minutes long.
They have ergonomic interference.
Figure 3 - Graph O.E.E. Cell divided into four classes - Source: Authors.
From the 08 cells that compose this group, 06 are high performance and 02 with average performance close to 85% and none with low performance. All three have in common:
Only one operator that interferes with the production. e.g.: the cell is composed one single employee loading the bottleneck machine.
Setup medium with time under 10 minutes.
The bottleneck is out of cell.
Figure 4 shows the graph of the dispersion of cells issued by software.
Figure 4 - Scatterplot of Characteristics of Four Classes
From the observed influence of the 16 main features production by cells in terms of relative participation in the group, along with the results of the three most significant features in the four classes according to the correspondence test, the research identified the most representative features and corresponding each pattern efficiency. The most significant matches are verified in the comparative table 7.1.
Table 7.1 - Summary of characteristics found in the manufacturing cells
The high-performance cells have in common the following characteristics:
Load/unload automatic, meaning: The operator has none or limited influence on the pace of work on the bottleneck;
One operator station, meaning the operator is working only in the bottleneck machine/station, not working in two or more machines/stations within the workplace.
Workstation does not have any ergonomic constraint.
Preventive maintenance is applied
Type of machined material - steel.
Average time spent on each set up until 10 minutes.
Number of weekly setup equal to one or zero.
The lower the number of the operators in the manufacturing cell, the better is the performance
Some recommendations emerge from the data with the objective of achieving high performance in most cells and observed using tools available in the lean production model:
Standardized work of operators prioritizing bottleneck machines cells. Standardized work well applied does not require the bottleneck machine is removed from the cell and ensures their load as observed in the data that point to high performance machines with single operator (characteristics a, b, h). If standardized work is applied carefully about leveling tasks between operators avoiding idleness, will also prevent interference with the performance of the machines caused by lack of synchronization in larger groups of operators (characteristic h)
Assuming that the standardized work well distributed and applied not bring the desired effects, yet there would be no need for full automation with greater investment. The data indicate that a localized automation in loading and unloading decreases the influence of rhythm operator and increases the performance class of the cell (characteristic a)
Use of the tool fast set-up or SMED should be intensified and associated studies for logic sequencing of change tools. The goal is to achieve set up below 10 minutes ( characteristic f, g)
The use of TPM focused on autonomous and preventive maintenance accounts for loss reduction with unscheduled downtime and helps the involvement of operators. (characteristic d). On machines used in working with harder materials like the example of cast iron, one additional preventive action should be considered or planned preparation to support these materials. (characteristic e).
Implement programs to improve the ergonomics of jobs is an initiative that has been adopted in kaizens to support standardized work (characteristic c).
The research has the limitation of not quantify the individual influence of each of the influencing factors. This limitation prevents to perform the prediction of how each recommendation for improvement in the factors mentioned could result in improvement. However, the triangulation of data to suggest that the gradual adoption of the factors to improve standardized work and to improve the availability of machines through TPM and fast set-up showed strong evidence in the facts and adherence to the theory.
These findings are of great relevance to emerging countries because it means that in a stage of intermediate technology, improvements in machine performance and productivity in general can still be achieved without large investments by using the recommendations and practices of production management compliant with the model lean production.