2010 Issue

37 Figure 2 – DOCS II vs. Method 9 continued on page 42 Conventional wisdom indicated that a certification process should be required. Because of the existence of a commercial infrastructure which supports human certification to the EPA’s satisfaction, it made sense to use what was already available and certified. Thus the ASTM standard would have the same certification requirements as a human observer, with some caveats. Yes, Certify . . . but How Often? Another critical consideration was the certification period. In theory, once a computer, camera model, and software combination is certified, it should require no further certification. The reality is, computers and cameras wear-out, and software changes. Again, frequency of certifica- tion is critical, as this issue drives recurring cost. Since both the camera and computer are electro-mechanical devices, their failures will follow a “bathtub” curve. Specifically, there will be infant mortality related failures (warranty issues), and then life-span and use-based failures. Based on the assessment of associated reliability and failure data, it was concluded that re-certificationmust be forced every three years. In addition re-certification must occur upon significant software changes as defined in IEEE 12207 Software Lifecycle Management (i.e. changes to the functions that enable analysis of the digital imagery). Humans Still Involved? Another critical decision was operator training. The first is ensuring the basic understanding of the practice of gathering opacity observations. As mentioned earlier, Method 9 field protocols are solid, so they were retained. However, they are detailed and rigorous, requiring some level of training through smoke school. The commercial market addresses this need to the satisfaction of the EPA. Thus the choice was made to incorporate this field protocol training with the certification requirement. The second component is tool-specific training which must be performed by software vendors. It is the expectation that commercial vendors will provide tools to meet this standard, and those who purchase these tools would be trained on their specific operation. Therefore, effectiveness of the training is reflected in the user’s ability to pass smoke school with their Method 9 ASTM Pre-Requisites *Smoke School Classroom Training (Once) *Same *Smoke School Field Evaluation (Every 6 Months) *Smoke School Field Evaluation (Every 3 Years) Defensibility *Historically Difficult To Defend *Easy to Defend *Forensic Support Consists of Forms and Human Recollection *Forensic Support Consists of Electronically Stored Media with an Image of the Event Field Protocol *Capture Location, Distances, Bearing, Weather, Sun Angle *Same (More Automated, Verifiable) *No Applicability Limits *Stacks >7ft. in Diameter Excluded Table 1 – Method 9 Vs. ASTM Standard