ABSTRACT

As mentioned in Chapter 1, it is extremely important that proper planning be done in first designing and then conducting experiments. The design engineer must adopt a proactive approach in setting up the design so that he/she is able to extract accurate information from the experiment. This ultimately leads to a product or service with minimal defects. On a new product, a well thought out design avoids rework during its development and implementation phase. It is with this motivation of objectivity in designs that we present some practical designs useful for the practicing engineer. This chapter presents a wide variety of design types. These include onefactor-at-a-time design, two-level full factorial designs, fractional factorial designs of resolution III, Plackett-Burman designs, fractional factorial designs of resolution IV, and fractional factorial designs of resolution V. Advantages and disadvantages of each design type are also discussed to enable the engineer in deciphering which design to use for a particular application. We begin with an example that shows how the one-factor-at-a-time design is very not efficient. Consequently, experimenters must use other types of designs that are cost efficient and easy to implement. An engineer is designing a battery for use in a device thatExample 2.1: will be subjected to extreme variations in temperature. The objective of the experiment is to maximize the lifetime of the battery. The engineer determined two factors that affect the lifetime of the battery: 1. Material type O N 2. Temperature 15°F 125°F Now that the engineer has determined the objective of the experiment, influencing factors, their region of interest, and the response, the next logical steps are to set up the design, determine the settings of the factors, and perform the experiment.