This article is published by China Science Software Network. If you have any questions, please contact us.
This case was provided by Andy Roff and Larry Blair, and their research was assisted by the model provided by the author. Both Andy and Larry are managers of auto parts and they own and manage several stores. Each of them has more than 30 years of experience, and they have a lot of experience in providing mutually beneficial information systems for suppliers and distributors. Their contact information is:
Background and history of the automotive parts market
A century ago, the auto parts market (AAM) was created since the car was not used. This may happen because the person who used to drive the carriage rudely uses the clutch. Therefore, when people need to replace the first screw, AAM is produced. As automakers increase, auto parts manufacturers are also beginning to increase (some of which are commissioned by automakers) to produce parts of various sizes, qualities and durability. At the time of this case study, World AAM sales were approximately $800 billion and the expected annual growth rate was 3%.
Because there are so many parts and suppliers, we need to compile them into a catalog. The use of single-chip miniature film in the 1960s was a major breakthrough, and it was almost exclusively used in the after-sales service network of automobiles. For automotive mechanics, the quasi-electronic database improves efficiency in parts sales and car repair processes. However, due to the high degree of competition in AAM, the use of a large number of expensive photo libraries has become increasingly infeasible.
In the 1980s, with the popularity of personal computers, they naturally became tools for compiling electronic catalogs. A major difficulty at this point is that different systems have different data formats, and printing these different formats can be costly. Similarly, when suppliers each need a parts assembly platform, they each design and install an electronic catalog. To make matters worse, these platforms are not compatible with the point-of-sale systems (PoS) introduced in the 1970s.
In terms of concept and technology, this demand gap is extremely huge. The Europeans' first attempt was to focus on providing stand-alone terminals that route parts from multiple suppliers in a "bookcase" format that allows the car to obtain information from suppliers and agents. The system originated in the Netherlands and was licensed in the UK in the 1990s.
Although the system ultimately failed, it allowed manufacturers and suppliers to focus on providing electronic data rather than printed data. This shift is reinforced by the ambitions of the nation's parts distribution chain, which provides electronic catalogs and requires the PoS system to bring electronic catalogs to every terminal. The behavior of producers and suppliers increases the demand for electronic data.
In the United States, national PoS suppliers decided to make large-scale investments in 1984, which allowed the United States to have a dedicated, compatible electronic catalog in 1985. The European electronic catalogue was developed five years later.
So manufacturers have set up third-party electronic catalog systems on the main facade. They were forced to reduce the use of data, reduce the time and precise control over the data submitted to the market, and they had to provide different versions of the electronic catalog for supply chains and third-party vendors in different countries. In some cases, they are even responsible for the configuration and display of the system.
Problems facing the entire industry
Every family has a difficult experience. As data becomes more versatile and meets industry standards (in the US, a list of industry sponsors is available), data is becoming more compatible and faster to market. Now that technology is becoming more advanced, the pictures and annotations of the parts, the location of the car, the precautions, and other key information need to be added to the catalog. These improvements can increase purchase experience and allow manufacturers to differentiate their products.
In addition, the products produced by the manufacturer will be accurately linked to the list of cars, which in turn will be linked to national statistics. Now, for the production and sales of automobiles, we can introduce risk analysis, simulation, forecasting, optimization and real option models. When considering the various possible impacts of manufacturing and searching for parts, we can see the vulnerabilities and vulnerabilities of traditional decision-making methods.
Analysis complexity
An example of this case study is based on Cashy Motors, a theoretically proprietary company that designs and manufactures automotive parts for original equipment manufacturers (OEMs) in the automotive industry. Casky specializes in the production of circuit breakers, which are often referred to as alternator starters. The company has close links with Ford and General Motors (GM), and these companies provide the foundation for Caky's operations in North America and Europe. As the development partner of the world's two largest automakers, Caky uses its engineering expertise to provide the two manufacturers with the world's most popular model launchers. Casky's partnership with the two companies has led to orders for new models that have maximized fuel efficiency. These models put higher demands on the starter, which in turn increases the cost and complexity. Casky won a contract for the production of the Phalynx minivan starter for GM in 2005. GM requires these starters to be installed in the car during assembly. However, Caky also won a contract to provide after-sales service for the starter. Considering economic value and employment rate, the automotive industry is the largest industry in these countries. Of course, North America and Europe have the largest number of cars in the world, and they also have the largest number of cars per capita.
Phalanx's sales in the first year is expected to reach 100,000 units, which will rise to 150,000 units in the second year and will be reduced to 100,000 units in the third year. The estimated sales volume for similar models is within 5%. The model will be produced in continental Europe and sold in North America and Europe. In the end, the number of models will vary between North America and Europe, accounting for 45% and 55% respectively. Different suppliers will provide statistics on the number of cars each year. The new model has a marketing life of 4 years, and the sales volume will drop steadily from 150,000 units to 75,000 units in the 5th year, just before the launch of another new model. Therefore, the total number of vehicles will reach 575,000, of which the annual scrap rate is 2% (this includes the full payment of insurance and the economical view that it should not be repaired). There are two types of gasoline and one diesel. The three engines are expected to be the same. They will be used on both original and refurbished models, but will not be used on all new models.
Casky was selected as the supplier of three engine starters. It only offers brand new (not repaired) products to GM and AAM. Each starter is different, they are uniquely designed and have different wear characteristics (MTBF minimum mileage before scrapping), the smaller gasoline engine is 100,000 miles and the larger gasoline engine is 85,000 miles. The diesel engine is 100,000 miles. The expected average mileage for the year is as follows: a smaller gasoline engine is 12,000 miles, and a larger gasoline engine and diesel engine is 15,000 miles. On Continental Europe, the warranty period is 2 years; in the UK, Ireland and North America it is 3 years.
Since GM has enough stock, it requires a 100% installation success rate in a week. The probability of failure (ie, the maintenance requirement before the MTBF is reached) is 1:10000. GM's retail business network has 250 facades in Europe and 150 facades in North America. Before the new models go offline, each facade must have at least two models. Casky has three European and two North American distribution points that serve GM's retail network and implement independent sales and retail chains for AAM. GM has the least number of vacancies, 20% for national chain stores and 25% for independent dealerships.
Caskey expects to provide GM with 90% of after-sales service (outside the warranty) and expects to compete with other manufacturers in the fourth year and enter the new regulator in the fifth year. European producers have a distribution network in North America, and the distribution network-modified starter is equally applicable to another model (this model has similar ownership and engine). The manufacturer expects the new model to gain a 10% market share from the very beginning, with an annual growth rate of 2% (double growth rate) and an additional model growth rate of 50%. This model is the original equipment of two auto parts. One of two manufacturers (VPMs). Two replacement devices are sold in North America and three are sold in Europe, each with an expected market share of 5% and only for AAM distribution. The service capability of the reset device is directly related to revenue (a total of 5 years, in the first year of operation, GM is 100%, and gradually declines with the decline of the new VPM in subsequent years). The MTBF of the reset device is only 66% of the new device.
Apply risk analysis, simulation, forecasting, and optimized analysis frameworks
Understanding and solving problems is not an easy task, and you need to be proficient in using Monte Carlo simulation, prediction and optimization of Risk Simulator®. Figure 1.1 shows a model of vehicle demand based on the assumptions previously listed. The maximum, minimum, and value range are listed. Figure 1.2 shows the simulation results for the demand for each cycle. Simulation results are available every quarter in Europe and the United States, which corresponds to the annual values ​​(100, 150, 100, 150, 75000).
Figure 1.1 Car demand forecast
Figure 1.2 Monte Carlo simulation of vehicle demand forecasting
Figure 1.3 shows the other constraints and requirements of the model, such as the scrap rate of each part of the car, the wear rate and the average number of miles traveled per year. We should note that the cells highlighted in Figures 1.1 and 1.3 are simulations of constraints, each of which is the result of thousands of iterations in the model. Next, based on the uncertainty of these demand, we will use the optimization model (Figure 1.4). In this model, the decision variable is the quality of production (ie, given the expected demand based on uncertainty, we need to find the optimal production quality). In this model, we consider the price per unit, the scrap rate, and the average mileage per year. This analysis provides the best quality of production, so the total net profit can be maximized, thus eliminating the extra cost of holding too much or too little inventory.
Figure 1.3 Additional requirements
Figure 1.4 Optimization Model
For example, suppose the cost of holding one more unit is $1.00, and the loss of sales holding less than one unit is $1.20. In addition, the minimum holdings for inventory are 800 in 6 months, and the minimum sales in the world is 400. Finally, the annual production volume cannot exceed 1.5 times the expected production volume, so that the market can be kept stable. Monte Carlo simulation and prediction methods are also applied to dynamic optimization. The true quality of the parts produced by the manufacturer can maximize net profit and minimize additional losses, and they are always affected by the maximum and minimum values ​​of the parts produced by the manufacturer (Figure 1.4, Figure 1.5). As can be seen, the optimal method is to produce a smaller quantity when the introduction of Phalanx is started; gradually, as the vehicle age increases, the number of parts is increased. The peak of the number appears between the seventh and tenth years, when the warranty expires and the components are most in short supply, and then the number gradually declines (the car will be decommissioned, sold or abandoned).
Figure 1.5 Optimal production and production constraints
Using these advanced analytics, we can predict optimal yields, the lifecycle of specific components (based on historical data and thousands of simulations of potential results in an optimization model). In fact, we can go one step further. After completing the optimization analysis, we can repeat the simulation and get the probability distribution of the net income of a certain part (Figure 1.6~Figure 1.8).
Figure 1.6 shows the net profit of a specific accessory over its lifetime, with a confidence interval of 90% and a distribution between $15.64 and $18.87 (million). We can conclude that if the current parts are manufactured, the probability that the net profit of the current parts will exceed the revenue of other businesses is 91.20%.
In contrast, if you do not use optimization and simulate risk, you will get a set of sub-optimal results. For example: assuming the average of the production forecasts (based on the maximum and minimum production required for each cycle), the total net profit would be $13.43 (million); if the production requires the minimum production, then the total net profit It will be .71 (million). Therefore, given the large fluctuations in the values, the operational optimization can guarantee a maximum net profit of $17.54 (million) based on uncertainty and risk.
We can summarize the above discussion as follows: Monte Carlo simulation, prediction, and optimization play a crucial role in determining risk components, pricing uncertainty, and demand. In addition, this analysis can give quantitative results to guide the producer's production. Therefore, with the help of risk analysis, decision makers can not only decide what to produce but also how much to produce; and also determine the optimal sales price after production, thereby maximizing profits and minimizing losses and risks.
Smoking Accessories,Glass Smoking Accessories,Wood Smoking Accessories,Cigarette Holder Parts
NANTONG OUFEIYA IMP& EXP CO.,LTD , https://www.nantongoufeiya.com