Modern steel mills are very automated and need to monitor their substantial energy costs carefully to be competitive. In making cold-rolled steel (as used in bodies of cars), it is known that temperature during rolling and the amount of expensive additives (expensive metals like manganese and nickel give steel desired properties) affect the number of pits per 20-foot section. A pit is a small flaw in the surface. To save on costs, a manager suggested the following plan for testing the results at various temperatures and amounts of additives.

90, 0.5% additive

95^o, 1.0% additive

100^o, 1.5% additive

105^o, 2.0% additive

110^o, 2.5% additive

Multiple sections of steel would be produced for each combination, with the number of pits computed.

(a) If Temperature and Additive are used as predictors together in a multiple regression, will this approach yield useful data?

(b) Would you stick to this plan, or can you offer an alternative that you think is better? What would that approach be?