Đang tải... (xem toàn văn)
Tài liệu: Giới thiệu về quy trình quản lý dịch vụ ô tô tại đại lý Toyota Trong thiên niên kỷ mới, ngành quản lý dịch vụ, vận hành đóng một vai trò quan trọng trong mọi doanh nhân, các ngành công nghiệp hiện đại hóa và các công ty tiên tiến trên toàn cầu. Một thực tế hiển nhiên là dịch vụ không chỉ hỗ trợ cho bất kỳ doanh nghiệp nào mà còn nâng cao và phát triển chất lượng của họ vì sự tiện ích của khách hàng.
TABLE OF CONTENT CHAPTER 1: INTRODUCTION TO THESIS 1.1 Overview 1.2 The object and scope of research 1.2.1 Object of Research 1.2.2 Scope of the Research 1.3 Research methodology 1.4 Outline of thesis CHAPTER 2: THEORETICAL BASIC 2.1 Introduction to operations management 2.1.1 Introduction 2.1.2 The Customer’s View of The World .3 2.1.3 A Firm’s Strategic Trade-Offs 2.1.4 Overcoming Inefficiencies: The Three System Inhibitors 2.1.5 Conclusion .6 2.2 Introduction to processes .6 2.2.1 Process Definition, Scope, and Flow Units 2.2.2 Three Key Process Metrics: Inventory, Flow Rate, and Flow Time .9 2.2.3 Little’s Law – Linking Process Metrics Together 10 2.2.4 Conclusion 11 2.3 Process Analysis 11 2.3.1 How to draw a Process Flow Diagram 11 2.3.2 Capacity for a One-Step process 16 2.3.3 How to Compute Flow Rate, Utilization, and Cycle Time 16 2.3.4 How to Analyze a Multistep Process and Locate the Bottleneck 18 2.3.5 The Time to Produce a Certain Quantity 18 2.3.6 Conclusion 19 2.4 Process Improvement 19 2.4.1 Introduction 19 2.4.2 Measures of Process Efficiency 20 2.4.3 How to Choose a Staffing Level to Meet Demand .22 2.4.4 Off-Loading the Bottleneck 23 2.4.5 Conclusion 23 2.5 Process analysis with multiple flow units 23 2.5.1 Introduction 23 2.5.2 Generalized Process Flow Patterns .24 2.5.3 Attrition Losses, Yields, and Scrap Rates .24 2.5.4 Flow Unit – Dependent Processing Time .25 2.5.5 Conclusion 28 CHAPTER 3: RESEARCH METHODS 28 3.1 Research in operations management 28 3.1.1 The aim and the scope of research 28 3.1.2 Roles of researcher .28 3.1.3 The research process 29 3.1.4 Research as contribution to knowledge 30 3.1.5 What to research for academia and practice? 31 3.1.6 Research quality 32 3.2 The research process 33 3.2.1 Contributing the knowledge 33 3.2.2 Using literature to develop the research topic .34 3.2.3 Considerations in choosing a research approach 35 3.3 Surveys 37 3.3.1 The survey research process 37 3.3.2 What is needed prior to survey research design? 37 3.3.3 Data Analysis and Interpretation of Results 39 3.3.4 Questions to Check the Quality of an Ongoing Survey Research 39 3.4 Case Research .41 3.4.1 Introduction 41 3.4.2 When to use case research 41 3.4.3 The Research Framework, Constructs and Questions 43 3.4.4 Choosing Cases 44 3.4.5 Developing Research Instruments and Protocols 45 3.4.6 Conducting the Field Research .47 3.4.7 Reliability and Validity in Case Research .50 3.4.8 Data documentation and coding 50 3.4.9 Analysis 51 3.4.10 Conclusion 52 3.5 Longitudinal field studies 52 3.5.1 Introduction to the longitudinal field study 52 3.5.2 Setting up the longitudinal field study 53 3.5.3 Collecting data in the longitudinal field study 53 3.5.4 Analyzing longitudinal field data 53 3.5.5 Building theory from longitudinal field studies 53 3.5.6 Evaluating theory from longitudinal field studies 54 3.6 Modelling and simulation 54 3.6.1 Introduction 54 3.6.2 Origins and development of model-based research in OM 54 3.6.3 Methodologies in quantitative modelling .55 3.6.4 Conclusion 57 CHAPTER 4: DATA ANALYTICS 58 4.1 Type of data we use .58 4.1.1 Categories of Descriptive Data .58 4.1.2 Measure of central tendency the mean median mode and their interpretations and calculations 59 4.1.3 Measure of spread in data, the range, interquartile – range, standard deviation and variance 61 4.2 Relationship between variables 62 4.2.1 The covariance and correlation measures .62 4.2.2 Probability and Random Variables; Discrete Versus Continuous Data, Probability Density Function and Area Under the Curve .64 4.2.3 The Normal Distribution (Bell Curve), Norm.Dist, Norm.Inv Functions in Excel 65 4.3 Distribution and probability 68 4.3.1 The binomial and Poisson distributions 68 CHAPTER 5: CONCLUSION AND IMPROVING SUGGESTIONS 71 5.1 Conclusion 71 5.1.1 Summarizing the data in separate sub-processes 71 5.1.2 Analyzing and Finding the Bottlenecks of the sub-processes .74 5.1.3 Detailed analysis of bottleneck process 75 5.2 Improving suggestions for Car Dealer 77 5.2.1 Process 77 5.2.2 Quality 80 LIST OF FIGURES Figure 2.1 Consumer Utility, its components and subcomponents Figure 2.2 Express Maintenance Process Figure 2.3 Quick Repair Process Figure 2.4 Toyota Bien Hoa Service Process Figure 2.5 Flow Unit of a Process Figure 2.6 General Process Flow Diagram .13 Figure 2.7 Activities at Reception sub-process of EM service .14 Figure 2.8 Complete Process Flow Diagram 16 Figure 3.1 Different process paths in case research 43 Figure 3.2 Choice of number and type of cases 44 Figure 3.3 Reliability and validity in case research 50 Figure 3.4 Research Model 56 Figure 4.1 Positive covariance and Negative covariance .63 Figure 4.2 Covariance formula Copyright® eduCBA 63 Figure 4.3 Example of Continuous Data (Copyright®: towarddatascience) 65 Figure 4.4 Example of Normal Distribution (Copyright®: alamy) 66 Figure 4.5 Bell curve properties of the distribution (Copyright®: alamy) .67 Figure 4.6 Poisson distribution .70 Figure 5.1 Histogram of Reception 71 Figure 5.2 Histogram of EM 72 Figure 5.3 Histogram of General Jobs 73 Figure 5.4 Histogram of Delivery 73 Figure 5.5 Required time of each sub-process .74 Figure 5.6 Time required for each sub-process 75 Figure 5.7 Time required for each task at Express Maintenance station 76 Figure 5.9 Time required for each task at General Jobs station 77 LIST OF TABLES Table 2.1 General Workflow Example 20 Table 2.2 Definitions of Measures 21 Table 2.3 Cycle Time of Sub-processes 22 Table 2.4 Demand for agents at 10-hour shift 26 Table 2.5 Calculating the time required 26 Table 2.6 Computing Demand of each Sub-process .27 Table 2.7 Implied Utilization of each sub-process 27 Table 4.1 Example of calculating Mean 60 Table 4.2 Example of Discrete Data .65 Table 5.1 Calculated data of each task 76 Table 5.2 Time-consuming of each task at GJ sub-process 77 ABSTRACT BP – Body and Paint EM – Express Maintenance GJ – General Jobs ODO – Odometer or Odograph OM – Operations Management PMF – Probability Mass Function STDEV – Standard Deviation CHAPTER 1: INTRODUCTION TO THESIS 1.1 OVERVIEW In the new millennium, operations management service plays a crucial role in every entrepreneurs, modernized industries and advanced companies all around the globe It is the obvious fact that service not only support any businesses but also enhance and develop their quality for customers’ utility Advanced operations mean teaching young readers the content they need in today’s world, not the world of 40 or 50 years ago Therefore, “services” and “global” are incorporated throughout, rather than confined to dedicated chapters Manufacturing cannot be ignored, but again, the emphasis is on contemporary problems that are accessible and relevant to young learners For instance, Toyota Bien Hoa station is important for the functioning production, but young generation no longer need to be able to replicate those calculations Instead, they should acknowledge how to identify the bottleneck in a process and use the ideas from the Toyota Production System to improve performance And readers should understand what contract manufacturing is and why it has grown so rapidly Summary, we want young readers to see how operations influence and explain their own experiences General view of operations means educating young readers much more than how to math problems Instead, the emphasis is on the explicit linkages between operations analytics and the strategies organizations use for success For instance, we want young generation to acknowledge how to manage inventory In other words, general view of operations supplies readers with a brand new, broader perspective into the organizations and markets they interact with every day We strongly believe that operations management is as relevant for a young adult’s future career as any other topic taught in a business school New companies and business models are created around concepts from operations management Established organizations live or die based on their ability to manage their resources to match their supply to their demand One cannot truly acknowledge how business works nowadays without understanding operations management To be a bit colloquial, this is “neat stuff,” and because the young will immediately see the importance of operations management, we hope and expect they will be engaged and excited to learn 1.2 THE OBJECT AND SCOPE OF RESEARCH 1.2.1 Object of Research - Data Analysis - Automotive Operation Management service theory - Research methods about collecting figures in sub-processes 1.2.2 Scope of the Research - Collecting and analyzing figures from three main sub-processes - Acknowledging the first five chapters from Operation Management material - Learning Data Analysis to apply on Automotive Service 1.3 RESEARCH METHODOLOGY Research Method is the ideal way to observe and collect data correctly We will demonstrate with detail in Research Method section in this graduated thesis 1.4 OUTLINE OF THESIS Chapter 1: Introduction to Thesis Chapter 2: Theoretical Basic Chapter 3: Research Method Chapter 4: Data Analytics Chapter 5: Conclusion and Improving Suggestions CHAPTER 2: THEORETICAL BASIC 2.1 INTRODUCTION TO OPERATIONS MANAGEMENT 2.1.1 Introduction Generally, in this very first phase, we would like to emphasize the main elementary difficulties of connecting supply to demand This beginning desire us to focus on demand, what purchasers really want? After the demand is acknowledged, we take a careful look at the aspect of a solid experiment to deliver the demand, we look at the supply procedure Following that, we consider the operational choices, a firm must accomplish to grant consumers with what they want at a minimum charge At the present, regularly, customers need better, more development and improvement in products for affordably cheap costs However, in the particle modern world, this happening maybe not often be elementary to settle Usually, a consecutive segment in this first phase discuss about overwhelming three main prevention that maintain the service from bringing outstanding merchandises at low expend In addition, these avoidances, the activity also requires creating trade-offs and equity multiple, promisingly competing objectives We come to an end in this introductory stage by clarifying what occupation suited to operations administration look alike and by supplying a short and solidified general view of service control 2.1.2 The Customer’s View of The World Figure 2.1 Consumer Utility, its components and subcomponents Utility – A certain amount of preference that buyers really want for any product or service Consumers buy the product or service that maximizes their utility Most customers are looking for maximum utility For instance, customers want to purchase a cheap, spacious, durable and reliable vehicles, which most of them are Toyota brand in Sigma from the mean, Sigma about the negative fade of the coordinate axis and the Sigma about the right fade of the coordinate axis NORM.DIST NORM.DIST(x,mean,standard_dev,cumulative) X (required) – The value for which you want the distribution Mean (required) – The arithmetic mean of the distribution/ Standard_dev (required) – The standard deviation of the distribution Cumulative (required) – A logical value that determines the form of the function If cumulative is TRUE, NORM.DIST returns the cumulative distribution function; if FALSE, it returns the probability density function Let's walk through an example to better understand NORM.DIST Receiving work at the input at the agent has mean = 16 minutes and Standard dev = 20 minutes Calculate the ability to receive customer longer than 30 minutes We write the formula for this request as follows: = 1-NORM.DIST (30,16,20, TRUE) = 0.241964 That means 24.19% will receive guest larger than 30 minutes The reason we have to subtract from the rest is because this finite equation always gives the value to the left of the number line, which means the probability that the job will take less than 30 minutes, so we have to take - NORM DIST () to find the probability of performing a job is over 30 minutes 4.3 DISTRIBUTION AND PROBABILITY 4.3.1 The binomial and Poisson distributions Binomial distribution and Poisson distribution are the two most common distribution types for discrete data types Binomial distribution Binomial distribution is the distribution of probability experiments whose output consists of only two output values for each test We use the BINOM.DIST function for independent tests and the success rate is constant throughout the testing process BINOM.DIST(number_s,trials,probability_s,cumulative) The BINOM.DIST function syntax has the following arguments: 66 Number_s (required) – The number of successes in trials Trials (required) – The number of independent trials Probability_s (required) – The probability of success on each trial Cumulative (required) – A logical value that determines the form of the function If cumulative is TRUE, then BINOM.DIST returns the cumulative distribution function, which is the probability that there are at most number_s successes; if FALSE, it returns the probability mass function, which is the probability that there are number_s successes Example of the BINOM.DIST function: calculate the probability that dice shakes to the dial out of 10 attempts Compared to the above formula we can write BINOM.DIST(4,10,1/6,FALSE) while: is the number of times we want to fine-tune the probability of a 10-dice roll success test, of which are dial 10 is the number of retries of the test 1/6 is the probability of success in attempt Because the dice has faces, the probability of appearing on each face in attempt is 1/6 FALSE is because we want to calculate the probability of occurrences of the dial 6, if we want to calculate the maximum of times the number 6, then change the value of the last attribute TRUE bar, then the formula our will be BINOM.DIST (4,10,1 / 6, TRUE) So how will the BINOM.DIST function be applied to real car dealerships? For example, the dealer had statistics and obtained data that 3% of cars repaired and serviced at the dealer would result in quality problems and the dealer was responsible for reimplementing These jobs are free for customers Therefore, the dealer wants to calculate whether to take 100 cars at random from maintenance and repair jobs, the probability of having up to vehicles having an error and having to repeat the process is what percentage We built the formula to calculate based on the BINOM.DIST function as follows: BINOM.DIST(6,100,0.03, TRUE) POISSON DISTRIBUTION Poisson distribution is a graph that indicates the distribution of a sequence of discrete data recorded at a random time and the length of each recorded equal For example, the 67 number of customers entering the dealer at fixed hours, the number of repaired cars that have problems, etc And the results of those records not interact with each other To calculate the Poisson distribution in Excel we use the function: POISSON.DIST(x,mean,cumulative) X (required) – The number of events Mean (required) – The expected numeric value Cumulative (required) – A logical value that determines the form of the probability distribution returned If cumulative is TRUE, POISSON.DIST returns the cumulative Poisson probability that the number of random events occurring will be between zero and x inclusive; if FALSE, it returns the Poisson probability mass function that the number of events occurring will be exactly x To better understand the Poisson distribution, let's take an example of the estimated number of vehicles that will fail after repair A car dealer recorded an average of cars a month after repairing faulty by the dealer Let's fine-tune the probability that only cars will have problems next month POISSON> DIST (2,3, FALSE) = 0.224042 (22% chance there will be repaired vehicles that have problems in the next month) If you want to know more about the probabilities of different number of defective vehicles, you can create a table and graph Poisson distribution from that table as follows Figure 4.18 Poisson distribution 68 According to the map, we see that with the test of or vehicles with errors, the probability of occurrence is the highest 69 CHAPTER 5: CONCLUSION SUGGESTIONS AND IMPROVING 5.1 CONCLUSION 5.1.1 Summarizing the data in separate sub-processes In this stage, we use histograms to describe the trend of the amount of time needed for the word sub-process, for instance, how many cases the customer is completed in 10-20 minutes And use the average value of a data set to compare with the other two subprocesses to find the process that causes bottleneck in the process Reception We have a histogram showing the reception time from the time the customer drives into the Toyota Bien Hoa company until the Service Advisor completes the task and create a repaired order to the customer Hist o g ram o f recepti o n 40 45 50 35 30 25 20 15 10 0 6 11 22 Frequency 51 Frequency 55 60 M ore amount of time in minute Figure 5.19 Histogram of Reception Histogram demonstrates that, there are cases received within to minutes, the most is the case of receiving within to 10 minutes with 51 cases, and decreasing to 22 cases The longer time is from 10 to 15 minutes The average value of the reception period is 16 (minutes) 70 Production Express Maintenance The time measurement for the production sub-process is defined as the time from the Coordinator calls the Technician to receive the repaired order until the task is completed, the Technician will return the repared order to the Coordinator In this service process, we divided into main production: Express Maintenance and General Jobs Although the names of the two production are different, the definition of time for collecting data is the same Frequency Histogram of EM 10 10 20 30 40 50 60 70 80 90 amount of times in minutes Figure 5.20 Histogram of EM Based on the chart, we can clearly see the majority of customers in the fast maintenance phase usually takes only 20 to 40 minutes, and the average value of this data set is 36.33 minutes General Jobs 71 Histogram for General Jobs Frequency Frequency 50 100 150 200 250 300 More amount of time Figure 5.21 Histogram of General Jobs According to the histogram, customers usually wait between 100 and 200 minutes for their vehicle to complete, the mean of this data set is 125 minutes Delivery Histogram of delivery 16 14 12 Frequency 10 Frequency 0 20 40 60 80 100 120 140 160 180 200 220 240 More amount of time in minute Figure 5.22 Histogram of Delivery 72 The graph illustrates that the time distribution for vehicle delivery to customers is usually between 40-60 minutes and the average value of this data set is 60 minutes The reason for the long delivery time for customers can come from main reasons: Due to insufficient Service Advisors during rush hour When there are a lot of input customers and the quantity of completed cars has been repaired, it will cause Service Advisors to overload At the delivery sub-process, we need two factors: the car is ready to deliver and the customer is ready to receive the car, lacking one of the elements on the delivery sub-process will be difficult 5.1.2 Analyzing and Finding the Bottlenecks of the sub-processes Required time 140 120 100 80 60 40 20 CUSTOMER GREETING EM GJ Delivery Figure 5.23 Required time of each sub-process After collecting the average value of the work, we proceed to model on the bar chart for easy observation We see the general service progress and cars that require General Jobs are the most time consuming However, initially we divided the service process into three small processes, including reception, services, and delivery The production step consists of two small workflows that are Express Maintenance and General Jobs so we will take the average of these two jobs as the time representing the production step 73 time required for each sub-process 90 80 70 60 50 40 30 20 10 CUSTOMER GREETING PRODUCTION DELIVERY Figure 5.24 Time required for each sub-process It is easy to know that the bottleneck of the process that takes place at the services step Based on the operation management theory, we can reduce the load on bottleneck stations by dividing the amount of work for less time-consuming processes However, for car dealers, this is not possible because the steps of reception, carrying out services and car delivery have different characteristics, so we not share the works and tasks for the remaining steps 5.1.3 Detailed analysis of bottleneck process After it has been identified where the process bottleneck is taking place We break down the production step into sequential sequences of work to find out what makes the whole process slow down In this section, I use the box plot chart because the box plot chart feature not only tells us the average value of the data set but also lets us know the spread of the data set Below is a box plot diagram of detailed time of the Express Maintenance station (EM) including tasks, receiving repaired orders, taking genuine parts, conducting the service and finally checking and paying repair order 74 Figure 5.25 Time required for each task at Express Maintenance station With detailed data of each task is presented below Table 5.10 Calculated data of each task It is clearly that the service execution step is of course the most time consuming step and also the step with the largest range of fluctuating data, from minutes to 82 minutes For the data range with such a large difference, I will take the data range in the range of 1st and 3rd quartiles, so the time for production at EM station will fluctuate in the range 18 minutes to 33.25 minutes And the average time for EM station to complete a process is 28.4 minutes Next, we conduct the time analysis of specific tasks of the GJ production 75 Figure 5.26 Time required for each task at General Jobs station The detailed data table of the above box plot chart is shown below, Table 5.11 Time-consuming of each task at GJ sub-process The service step is still the most time-consuming step in the process, and it is also the step with the largest fluctuation time range from 35 minutes to 129 minutes And the average takes 94 minutes to complete car 5.2 IMPROVING SUGGESTIONS FOR CAR DEALER 5.2.1 Process 5.2.1.1 The Process of Reception The general vehicle inspection by Service Advisors is insufficient + For instance: When the vehicle is performed any maintenance service, and during the vehicle inspection, the Service Advisor does not check the vehicle’s battery, due to the 76 fact that the vehicle’battery is still function but is essentially weak After a few days, the customer returns and reports an error again + Reasons: The list of Service Advisor's tasks is not as comprehensive and well enough as Toyota's general process + Solutions: Should add up more tasks for the Service Advisors (or another position that is not in the production content because this is currently a bottleneck that needs to be handled) Improperly scheduled customer’s car appointments + For instance: In general, the quantity of customers’ vehicles using services will be crowded on Monday, Tuesday, and Saturday of the week, and in the morning at dealer will have more cars than the afternoon of the day At such times, Service Advisors must simultaneously pick up and deliver the vehicles if there is no reasonable appointment to avoid overcrowding + Reasons: The objective reason is that the majority of repaired cars at dealers are from other businesses, so it is difficult to change the schedule of repairing and delivery of the car Subjective reason is due to the fact that the Service Advisors have not properly booked the appointment + Solutions: Service Advisors should make appointments based on changes in the distribution of vehicle arrivals in and out of dealers 5.2.1.2 The Process of Production Lack of old or outdated equipment, failing to meet current repair needs + For instance: - The manual cleaning, overhaul and refresh of machine parts will take a lot of time and human resources - Another typical example is the lack of a gearbox lifter (or some other specialized equipments) If there is more than one repairing job using a gearbox lifter in a same time period, there will be congestion in the repair progress For example, finding and waiting for other Technicians who are using the gearbox lifter will create waiting times (times that not generate profits for dealers) + Reasons: Dealer has not enough invested in facilities to meet the needs of work and improve quality + Solutions: 77 - Temporary solution: Attach the nameplate to the position of the gearbox lifter in the equipment compartments Thus, the Technician will save time for searching for it - Long-term solution: dealer should invest more in facilities to meet the needs of the work and improve the quality (cleaning liquid or a specialized washing machine for cleaning, repairing and renewing machine parts; gearbox lifter and other equipments) Difficulties in locating vehicles waiting to be repaired + For instance: There are many cases after receiving a repair order, the Technician is unable to locate the vehicle despite the assistance and instructions of the agent's staff (Service Advisor, receptionist, etc.) ) This will result in affecting the work flow, agent resources (parking space, human resources, ) + Reasons: There are a lot of cars but there is no proper vehicle arrangement leading to parking the cars improperly and moving the car in an disorderly manner From the moment the car is parked in the waiting compartment until the car is repaired, no employee is officially responsible for managing the location of the car park waiting to be repaired (according to Toyota's process standards, coordinator is responsible for this task) Moreover, Technicians, or even Service Advisors have not fully used the status label (after the period of observation during internship, this task is performed by staff in a non-professional manner) + Solutions: - Temporary solution: Arrange the vehicle appropriately Currently, compared with other dealers, Toyota Bien Hoa dealer is lacking a clear division of the waiting compartment (General Jobs, Express Maintenance, body and paint) On the other hand, every employee should pay more attention to placing and using status labels appropriately - Long-term solution: Redesign, rearrange the layout of the workshop, expand the space if necessary (by limiting resources that take up a lot of space but bring low productivity) Redesigning the status label to be more detailed so that it is easy to classify the current status of the vehicle, thereby reducing the time to determine the status of the vehicle, helping employees to reduce the time redundancy (time does not bring profit for dealer) 5.2.1.3 The Process of Delivery Current vehicle delivery process: when handing over the vehicle, the Service Advisor will hand over to the customers and then customers not receive the ticket to go out of the 78 gate and look for their cars → high time consuming According to the standard process, the Service Advisor goes to customers’ cars to hand over quickly to them Sometimes there is a overload problem at the repair process, the Service Advisor keeps track of the schedule work flow to make appointments that are not flexible, resulting in work jams 5.2.2 Quality The facilities to meet the needs of customers are not yet completed and satisfied + For example: Broken hot / cold water machine, the water and food serving menu are still very simple, which are small factors that contribute to customer dissatisfaction when returning to the dealer or making lunch stay decisions + Reasons: Dealer does not really care about the needs of the customer rest + Solutions: Need to invest more carefully on these basic issues, though small but they will change the trust of customers with dealer 79 REFERENCE 80 ... leads to a strategy for operations and for how the operations help the firm to compete in the market Operations systems are designed concurrently with the products and services that the operations. .. escalator, we try to count the quantity of shoppers riding with us, which is the inventory 11 metric for this process, but it is not easy to see everyone Besides, there seem to be too many customers... Odograph OM – Operations Management PMF – Probability Mass Function STDEV – Standard Deviation CHAPTER 1: INTRODUCTION TO THESIS 1.1 OVERVIEW In the new millennium, operations management service plays