The following are examples taken mainly from past students' class projects. Additional ones are taken from published sources, such as journals, magazines, and technical reports.

1. The cradle mount system, consisting of both rubber and metal assemblies, is used in automobiles to reduce the vibration and noise from the engine and from the ride. This will improve the ride characteristics of the automobile. In one study, the objective was to find an optimum configuration for the three variables: neck type, rate ring and material, that will give the best vertical dynamic rate. In this study these three variables were used at 2 levels each. The statistical tools used here are: Design of experiments and regression analysis. The student was able to collect the data and came up with a recommendation for the best design. In another study involving the cradle mount system, the student=s objective was to minimize the damping of a rubber component. Two factors: temperature and the time were considered at three levels. Using 3² factorial design, the student recommended an optimum setting for the temperature and the time.

2. This project involves the peening process, where in the operation involves moving aluminum material around a stainless steel ball. The ball is used to plug entrance holes where drilling operations are required. The objective of the study is to establish a planned replacement for the peen tools and to increase tool life and thus to increase machine uptime. Before the project was undertaken, there was no established tool life for the process and only visual inspection of the part was used to determine when a tool needs to be changed. He chose a set of four balls and used initial data to come up with some interesting observations. He indicates that this will be a starting point for further analysis involving other locations as well as using some additional factors in the study.

3. Paint process is a costly and a very involved one. Air spray pain guns and electrostatic rotary atomizers (referred to as bells) are two of the many ways to paint an automobile. Air spray guns use high air pressure to atomize and direct the flow of paint, and thus leads to a lot of wastage of the paint. In this project, a student was involved in finding optimal initial process parameters to achieve a target paint film thickness using "bells". The process parameters chosen for the study were: distance from bell to body, linear speed of the robot and bell, rotational speed of the turbine, paint flow rate, and shaping air pressure. Using design of experiments and regression analysis, the student identified optimal factor settings that resulted in film builds that were extremely close to the target value.

4. As is known Sulfur dioxide (SO₂) is one of the contributors to the "greenhouse effects". EPA is very strict about the amount of SO₂ released from the manufactruing processes. Many assembly plants and other component facilities produce the much needed steam through gas, oil or coal fired boilers. As is known, when burning coal, on average 98 % of the sulfur present in coal will be emitted as SO₂. The amount of SO₂ emitted is dependent only on the percentage of the sulfur in the coal, and not on the boiler size, the firing configuration, or the operation of the boiler. In this project, the student was interested in the amount of sulfur and the energy content of the coal supplied to her company. There were two coal suppliers to her company, one of which is believed to supply a better quality. She was interested in testing this claim as well as to determine whether one supplier has superior emission characteristics over the other, and if so, which parameter can be attributed to this phenomenon. Using statistical analysis (including 2-sample tests) the student came up with a number of interesting conclusions.

5. This project deals with evaluation of a prototype mechanism for fuel economy improvement. That is, a particular device, which is claimed to improve fuel economy by converting engine pumping loss into electrical output to supplant the engine driven alternator, and thereby reducing fuel consumption, was under study. The objective of this project was to determine whether the vehicle test data that were previously collected supports the need for the device. Using statistical tools (tests of hypotheses, regression analysis) the student concluded that the device does have an impact on the engine performance for combustion and emissions; and also the device could increase MPG.

6. This project deals with tube size variation study. Tube sizing is a process in which the diameter is changed for a given piece of tubing. This feature is critical in exhaust system applications. Tubes that are too large may not fit into mating tubes and thus prevent assembly. On the other hand, tubes that are too small will fit, but may not seal exhaust gases. Thus, it is very important to study the variation in the tube sizing process. After a brain storming session, the student identified 5 key variables (all at two levels) that might explain the variability and also help to find the best configuration setting. Using design of experiments, the student summarized his findings and recommendations for future work.

7. This deals with the problem of part thickness. A flat metallic part (which comes in varying thickness) is pressed to a specified thickness. Since parts were coming with various thickness initially, it is important to adjust the pressing mechanism to get the final product to be of the same thickness. The student used a linear regression model to predict the relationship between change in part thickness and the pressure applied.

8. Spark plug insulators are extremely abrasive and hence results in tool wear out prematurely. In this project, the student was interested in determining whether one or two different coatings will help extend the life of the nests that hold the parts. Using one machine and testing 3types of nests, it was determined that these three types do not significantly extend the life and it was recommended to test using different materials and coatings.

9. This project deals with window regulator pivot joint=s effect on the sash angle. The angle of the glass sash in the full up position is an important parameter and indicates how well the glass is sealing in the header. This is important as any leak will lead to structural problem. The student first developed a regression model of the current pivot design and then developed a similar one for an alternate pivot design. He concluded that the alternate design could control the variance better than the current one, but requires additional testing in view of an increase in the alternate design=s diameter.

10. In order to compare the system throughput before and after the initiation of process improvement efforts within the body shop of a leading automobile company, this student used standard statistical methodology to conclude that the process improvement team continue the efforts and also monitor the throughput to establish certain factors, such as cycle time and machine downtime.

11. Over the last few years the use of heated exhaust oxygen sensors has grown due to the emissions related benefits. With the increasing emphasis on quality improvements and process/product variation reduction, it has become very essential to look at and analyze process control data to ensure that key product characteristics are being met consistently, and the process performance is meeting or exceeding requirements. In this project, the student uses the data on ceramic heater resistance, which was compiled for a three-month time period, to make a number of useful recommendations and conclusions.

12. Testing electronic components to assure the resistive and capacitive values are correct for the components requires statistical analysis. In this project, the student was interested to see whether or not the test values obtained are consistent from tester to tester as well as from fixture to fixture within the testers. Using Analysis of variance techniques, some interesting conclusions and recommendations are drawn here.

13. In this project, the student=s interest was to verify whether the product HVM (heating and ventilating module) for cars, is of high quality and free from defects. He used basic statistical methods as well as his own program to analyze the data and make useful recommendations.

14. Color is one of the main ingredients in marketing strategy. We see these in food items, medicine, containers, packing materials, and so on. Consumers are attracted to color products more than plain ones. The product may not be good, but the presentation of the product may have more influence in people acquiring the product. One of the major problems in coloring products is environmental in nature. For example, in plastic industries heavy-metal pigment approach for coloring has been very popular until 1980's, when environmentally safer coloration began to emerge. One such method is using mixed-metal oxides (MMO's), which is quite efficient and more expensive than the traditional method. However, due to federal and state regulations, MMO's are being used. Statistics (DOE) is used to determine the optimal composition of MMO's.

15. Composite materials are used as pressure release systems in many devices. In order to improve the reliability assessment of these materials, statistics (regression analysis) is used.

16. Any company that produces a product invests time and money to study the product's reliability. That is, the company is interested to (a) know (predict) the lifetime of the product, (b) find how long a warranty, if any, should be given to the product, (c) calculate the probability that the product will function without any problem during the warranty period, (d) identify the causes of different types of failures of the product, (e) study alternative method, if any needed, that might improve the product, and so on. Problems of this nature fall under the general theme of "reliability data analysis" or "life time data analysis". Analyzing the reliability data is one of the major functioning components of the engineering and management department of the R&D of any company. Much of the analysis requires knowledge from probability and statistics.

17. Steel containers are used in transporting many goods including oil and nuclear power plant waste. Hence, it is very important to develop acceptance resistance criterion for the containers to brittle failures. The main variables of interest are the crack size and the stress intensity. Since these are random variables, the study of these requires probability theory and statistics.

18. Vinyls are used as siding in the houses to protect the brick walls (or dry walls). Since colored PVC can reach temperatures high enough to cause early failure of the vinyl siding. Hence, it is imperative to know the maximum temperature that a piece of pigmented PVC can reach when exposed to sunlight. Statistics (regression analysis) is used to predict the heat buildup of pigmented PVC as a function of the amount of sunlight absorbed by the PVC panels.

19. PCR resin is used in the manufacture of trash bags/liners. From environmental point of view it is very important the bags/liners have very high recyclable contents. This requires the study of the physical properties of the bags/liners that have different grade of polyethylene films and different resins. Again, DOE is used to determine the maximum performance characteristics.

20. Thin films are used in many diverse fields such as food packaging industries, housing industries, paper industries, and medicine. The film obtained from the blown film is known to be influenced by many variables. For example, in the polymer the variables are MW, MWD, density and branching; in the equipment, the die size, the die gap, the air ring, IBC, and IRIS are some of the variables; the output, melting temperature, and frost line height will be the variables in the process. The DOE is used to determine the effects of fabrication variables: the blown-up ratio (bubble radius/die radius), the specific output rate, the melting temperature, and the frost line height.

21. Cycle time (time between the completion of the last product and the completion of the next product) is an important quantity in the manufacturing processes. Depending on the product and the layout of the manufacturing area (some products such as a car and a semiconductor chip may require many different processing steps before rolling out as a final product)the cycle time, in reality, is a difficult one to quantify. The manufacturing company would like to estimate this variable as accurately as possible, to figure out the cost, isolate any problem, identify source, if any, of variability and so on. Probability and statistics are the essential tools for such study.

22. Package delivery companies are interested in scheduling the delivery trucks and the drivers to pick up and deliver packages. The companies have several hubs (central processing places) where the packages are sorted. One of the important factors in efficient delivery of packages is to estimate the number of drivers needed in each of the hubs for each of the working days as well as the season of the year. Regressions analysis and time-series analysis are used in developing statistical models for predicting the number of drivers as well as the number of packages. Also from the past data, ANOVA techniques are used to allocate the resources among the various divisions in each hub.

23. Devices such as the pillars supporting the platform of an offshore drilling unit, the pillars of a bridge, electronic items mounted on outboard the ship, and others, are exposed to several hazards in a marine environment. These hazards in the form of chemical, electrical, thermal, or mechanical stresses lead to degradation and eventual failure of the device. Maintenance and replacement of such devices are time-consuming and expensive. Hence, it is imperative to estimate the reliability of such devices as functions of time.

24. In order to determine whether there is any statistical difference between metal and ceramic substrates on the emission performance of automotive exhaust systems, a design of experiment was conducted. A fractional factorial design (2^5-1 design) was used to conclude that metal substrate may only be marginally better than ceramic substrate.

25. The assembly of a blower motor into a fan coil blower requires that a band be assembled around the motor, which captures and pins the assembly legs against the motor. The band is fastened using a bolt and nut through the ears of the band. It is noted that if the bold and nut assembly is over torqued, the band, which is fabricated from sheet metal, fractures in the bend forming the ears. This leads to eventual band failure resulting in more damage to the blower, its housing and the evaporator coil. Hence, a design of experiment was conducted to identify the optimum settings for nut material, bolt material, lubricant/no lubricant, and, nut and bolt size.

26. Rubber weather strips are used on automobiles to seal out water, dust, and vehicle and road noise. The seal around a vehicle door are some of the most important ones as the weather strips are easily noticeable by the occupants’ of the vehicle. During the introduction of a new model car, an unacceptable noise level was noticed and a team was assembled to resolve this problem. The team identified four key variables (automotive coating type, thickness, degree of cure, and weather strip (rubber) type. Using a 2⁴ design, optimum settings were identified and the production process used these settings to reduce the noise level.

27. An engineer in a small production process within the manufacturing plant was interested in comparing two 10.5-hour shifts with respect to the production output. These two shifts meet the same daily production requirements. In order to determine whether one shift cost less than the second one to operate (using electrical power consumption as the differentiating factor), the engineer collected data on a number of key variables. Using ANOVA, interesting conclusions were drawn.

28. An ABS manufacturing company was asked by one of its customers to reduce the base brake leakage in their ABS braking system solenoid values. To accomplish this, the engineer set up a DOE with many key factors and came up with very interesting findings.

29. A high percentage of "pole height" rejects was a concern of an engineer working in a manufacturing plant of sensors. A team consisting of 9 members identified key factors using fishbone diagram and conducted a series of experiments iteratively. The problem (at least at the time of submitting the report) was not solved. However, the conclusion of this report underscores the importance of applied statistics and so is quoted directly from the student’s report as follows. "It was a very good thing for me to be working in this project, because I was able to apply what I saw in the class and really understand the results and the steps we had to follow to determine the solution. It is very easy in the production real world that when you have a problem you start to shoot everything you think can be affecting, and sometimes you spend more time on this because you do not have the right tools of you just do not understand them. The problem has not been solved yet, but the variability was brought down to two major reasons and thanks to all the statistical tools that each member of the team knew how to use."

30. The primary function and design of the core crimp is to optimize the electrical performance of he cable/terminal interface for the service life of the terminal. The core crimp design utilizes the set of relationships between the terminal, the cable, the core crimp subtools, and the crimp die to provide this high quality connection needed for sufficient power and signal distribution in automotive applications. For signal distribution, or low energy systems operating under 5 volts, the core crimp connection must sustain optimal electrical performance in the face of extreme environmental conditions. This investigation uses regression analysis to establish a linear relationship between environmental conditions and core crimp geometry and their effects on core crimp resistance.

31. In this project dealing with the study of an automobile glove box door alignment, the student uses a 2⁴ design to identify the optimal settings that improves the door parallelism to measure within a given tolerance limit.

32. A multiple linear regression analysis was used to determine the effects of the duration of laser heat and the tension force on the taper length of an optical fiber