整合性火災風險評估分析在通信行動電話廠之應用與探討

Abstract

本研究係針對一通信行動電話廠進行完整之整合性火災風險評估分析。此研究從四個方向著手，第一為風險評估技術之建立，利用危害樹分析及作業安全分析與應用之風險評估軟體(RAMS)，針對廠內各項作業進行風險評估、控制與追蹤；第二為企業火災保險風險分析，係參考國際知名再保公司火災保險相關之風險評估方法，進行火災等級的分析，並尋找高火災風險製程單元與區域；第三為火災後果量化分析，利用NIST發展之場模式FDS來模擬高火災風險製程單元與區域所造成之溫度、煙濃度、殘氧濃度與二氧化碳之危害特性；之後為人員安全分析，利用IES所發展之人員避難逃生軟體分析人員避難模式與時間。最後歸納整體風險分析結果以了解通信產業之火災風險程度。 第一部份的評估結果發現，錫爐作業更換松香液體外洩造成火災與Burn-in Room高溫作業可能導致火災之發生，危害最為嚴重，應重新審核其標準作業操作程序之有效性及加強其改善預防措施。第二部份的研究顯示，高火災風險製程單元主要集中於使用易燃性化學品之機台（如錫爐製程區），而高火災風險區域則與高單價元件儲存區與成品儲存區有關。第三部份的分析發現，如增設排煙與灑水等防護設施，其火災危害相對其他較低；單獨設置排煙系統對於煙濃度控制最佳，但對溫度的影響有限；若增設撒水系統對於火場溫度具有控制(Fire Control)的效能，但是抑制(Suppression)效能並未顯現；另外亦發現火場主要危害為煙流危害效應。第四部份的分析顯示，對於大空間低人員密度區域，其人員避難快速，只有在靠近結點處方有累積現象產生，但相較於高密度人員區域，樓梯前與樓梯間則產生擁塞現象。此外，倘若將結果與廠內緊急應變紀錄相比對可以發現，若員工受過消防緊急應變避難逃生演練，其所花費之反應時間將較電腦模擬出之時間為短。 綜合以上結果，本研究探討如何以預測災害之定量模式後果模擬，取代並改善定性模式之評估結果，如此對於災害之掌握將更為明確，且可進一步評估防災系統之有效性(Effectiveness)及可靠性(Reliability)；此外，加強人員之應變演練，可縮短反應時間，減少人員之損失。

This research evaluates effectiveness and reliability of fire extinguishing systems by using renovating technological methods of risk assessment at a mobile phone manufacturing facility. The emphases of this study are focused on four stages. First, hazard tree analysis and RAMS are applied in the risk assessments of all tasks to evaluate the high-risk ones at this facility. Secondly, we use fire risk analysis techniques from insurance companies to identify the process units and areas with high risk potentials. Then, a fire simulation is carried out by using a field model, Fire Dynamics Simulator (FDS), developed by National Institute of Standards and Technology (NIST), to obtain the hazardous characteristics, such as temperatures, smoke generation rates, oxygen and carbon dioxide concentrations in the neighborhoods of process units and areas with highly potential risks. The last one is to apply the evaluation model, Simulex, developed by Integrated Environmental Solutions (IES), to analyze the employees’ evacuation time subjected to the dangerous conditions given by the previous stages. Finally, we can integrate the whole results of risk analyses to understand the fire risk in this mobile phone manufacturing facility. The results of part one show that the tasks of solder machines and burn-in rooms have the highest risks to cause the fire occurrences. Their procedure standards should be re-evaluated to avoid the fire situation. In the second part, it can be found that the process units, which use the flammable liquids, appear to have the greatest fire risk. At the areas with similar high risks, the hazardous levels are closely related with the numbers of workers, the frequency of operation and high-value equipment. Besides, the areas equipped with desmoke and automatic sprinkler systems have the relatively lower fire risk comparing with those without these systems. Establishment of desmoke systems is the best way to control smoke generation rate, but it has limited effect on temperature in fire fields. The automatic sprinkler system can enhance the control of fire, but it is not effective on fire suppression. The analysis of the last stage indicates that the less employee population in vast space, the better performances of evacuation, except some minor crowded movements occurred at the doors. However, the response time of employees, who had emergency response training before, would be shorter than the estimated time from computer system. Summarizing the results of the above-mentioned analyses under hazardous condition, this research can evaluate the effectiveness and reliability of fire protection system by using the quantitative models instead of the qualitative ones. Practices and emergency response training would lead to reduction of response time, and they can improve the overall safety level of the whole facility to avoid the losses of employees. Consequently, the control of fire incident would be improved.