EnergySafe Magazine Summer 2015, issue 38 | Page 20

20 Gas news Drilling down to the facts: improving gas incident reporting By Michael Weber, Technical Communicator ESV’s Gas Installations and Appliance Safety (GIAS) division has refined its approach to gas incident reporting after learning valuable lessons from our counterparts in New Zealand. In my last article I described the recording and reporting of gas incide nt data by Energy Safety New Zealand (ESNZ) for its quarterly and annual reports. I outlined the similarities and differences between our organisations and described how GIAS now plots gas incident data on maps and uses this information to identify clusters and trends, with expectations that this information may help ESV target problem areas with specific safety programs. In this article I will discuss the progress GIAS has made with data collection, data manipulation and report development. We are now able to drill down on incidents by gas type, environment, appliance type, incident cause and injury severity, to name a few categories. Gleaning the facts Much information is locked within each incident report statement received by ESV, and little information comes to the surface unless the statement is read in detail. Now report statements are being analysed and broken down into gas incident components and that information is stored for later use. As data populates the spreadsheet, pivot tables are employed to manipulate gas incident data into a number of tables and charts that show, at a glance, what is occurring. Gas incident data recording, in this detailed form, is still in its early stages; more data needs to be assembled, but we are on the way to providing a better view of the gas incident world in Victoria. To better understand the richness of gas incident data, incident reports are broken down into the following categories (similar to that employed by ESNZ): »» gas type »» environment »» appliance type »» cause »» incident type »» consequence »» injury type »» property damage »» incident notification. Each of these categories is further broken down into individual components. For example, appliance type is described as: »» cooker, oven »» heater »» hot water service »» barbecue 20 Residential incidents by equipment type Sample data for Jul–Sep 2014 15 10 9 6 5 4 2 1 0 Cook Heat HWS Inst BBQ Frid Light LPG Unkn A clearer picture: Greater interrogation of gas incident reports is helping ESV gain a better idea of what is causing gas incidents around Victoria. »» refrigerator »» light »» LPG storage cylinders »» unknown. A new style of reporting With additional information, now much easier to access, a change to the style of reporting has been developed. As shown above, the majority of incidents for this period involved heaters (as would be expected during the colder months of the year) and the breakdown shows there was one gas leak, one overheating issue, seven combustion issues, two fires and four incidents for unknown reasons. Separate from heater incidents, there were also nine installation incidents, which were all related to gas leaks. In addition to gas incident information, telephone Helpline calls are also recorded and presented by call frequency, call origin and contact method. This is useful for identifying the audiences of greatest need. Sample data from July to September identified that 61 per cent of calls received came from plumbers/gasfitters, with a further 30 per cent of calls coming from public inquiries. All gas incidents have a specific location, as identified by their address. This allows their locations to be plotted on a map. Data is plotted by layers so that incidents by gas type and appliance type can be shown individually or together. Incident clusters are easily identified, thus illuminating problem areas. Making the data work for us Greater interrogation of gas incident reports now provides additional detail and helps our understanding of the mechanisms around gas incidents. New data reporting methods provide information at a glance and with the advancements in data interrogation and improvements in reporting GIAS will achieve: »» quicker turnaround for statistical data »» greater detail of statistical information with the ability to drill down to specific event types »» greater ability to cross reference data, identify trends and make predictions. One further step that is being considered is that of chart and table data being directly plotted on maps, creating a 3D rendering of gas incident information. That way location and severity of incidents or types of incidents can be seen in one view; this becomes a powerful planning tool.   I f you have a query you can phone the Gas Technical Helpline on 1800 652 563