8.1 INTRODUCTION

Once the field measurements have been made, the data recorded and checked, downloaded, the entries made into our selected audit sheet or model, and the draft results output, we need to consider the validity of the results.

As with any audit in any industry, we need to make a certain number of estimations. These estimations may be about the accuracy of field data, missing system data, or the coefficients we use in our models. Estimations are usually made using some or all of the following:

• Modeled estimations

• Estimations made from good operator knowledge

• Estimations made from common sense

• Estimations made from data sets recorded in other areas or systems

As we go through the auditing, modeling, and estimating process we must record all assumptions and estimations made and the reason for making them. When we have finished the audit we must select the key criteria (the ones that will make the most substantial difference to the results) to check.

8.2 CONFIDENCE FACTORS

Confidence factors can be assigned to measurements made and to estimated data. These confidence factors can then be used to assign a bandwidth of possible real results on either side of the estimated result. Confidence factors can be assigned either as a result of detailed statistical calculations or based on operator gut feel. Either is valid as long as they are recorded with the reasons for assignment.

Figure 8.1: Sample sensitivity calculation. ( Source: Julian Thornton.)

8.3 SENSITIVITY...