In this case your bias is in favor of a "Cannot Reject the Null" which stands for "No statistical difference in the means" So run the following test and determine if the parts that are being manufactured are within specifications

Question: I used to manage a manufacturing plant making electrical switches. We molded 15 million polypropylene parts that went into the switches and one of the machines made a part that had a critical dimension ( a diameter on the shaft ) that we ran these tests to ensure they were made to spec. 
The population mean = 0.500" with a standard deviation of 0.003. 
Here is a random sample from one of the mold machines. The sample size is 30 and the average diameter is 0.4995" 
At the 0.01 alpha one tail test, determine if this lot is within spec or not