화학공장 사고예방을 위한 앗차사고 관리 및 사고 유발 요인 모델에 관한 연구

Presentation on theme: "화학공장 사고예방을 위한 앗차사고 관리 및 사고 유발 요인 모델에 관한 연구"— Presentation transcript:

1 화학공장 사고예방을 위한 앗차사고 관리 및 사고 유발 요인 모델에 관한 연구
Thank you chairman My name is Hyungjoon Yoon at Yonsei University in Korea. Coauthors are Mr. Hanyong Lee, Mr. Hyuckmyun Kwon and prof. Il Moon. Today I’m going to talk about near misses and their management to prevent accidents in chemical processing systems. This work is more relavant to industyr rather than academia. We gathered 2 thousands accident data for 6 months and I will show them to you here.

2 목차 1. 정의 2. 이론 3. 앗차사고 관리 필요성 4. 앗차사고 분석사례 (A공장) 5. 앗차사고관리 시스템 내용
6. 사고 유발요인 모델 필요성 7. 사고 유발 요인 관리 모델 8. 결론 Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

3 연구 목적 화학공장 사고 예방 기법 구축 앗차사고 관리 프로그램 개발 사고 요인 모델 개발
Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

4 산업재해의 크기 전세계 발생 건수 : 2억 5천만건 / 년 사 망 : 110만명 / 년 한국
발생 건수 : 2억 5천만건 / 년 사 망 : 110만명 / 년 한국 발생 건수 : 66,770 건 (1997년) 사 망 : 2,742 명 (1997년) 경제적손실액 : 7조 2천억 (1997년) Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

5 해외 중대산업사고 (예) Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

6 국내 중대 산업사고 (예) Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

7 1. 용어의 정의 1) 앗차 사고 부상등 피해가 없었으나 충분히 사고나 재해로 이어질 가능성이 있는 평범하지 않은 사건
2) 경재해 1~3일 정도의 작업 손실을 가져오는 사고 3) 중대 재해 4일 이상의 작업손실을 가져올 수 있는 사고 4) 사고 유발 요인 작업환경에 영향을 주어 사고로 연결 될 수 있는 위험 Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

8 연구 개략도 앗차사고 관리프로그램 사고유발요인 분석 프로그램 실제현장의 앗차사고 데이터 위험요인 발굴 (설문조사) Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident. 하인리히 법칙에 근거 사고유발요인모델 화학공장의 사고 예방

9 2. 이론 1 1 1 0 2 9 3 0 3 0 0 6 0 0 2) 버드의 법칙 1) 하인리히의 법칙 심각한 피해 중대재해
3 0 1 2 9 심각한 피해 중대재해 심각하지 않은 피해 경재해 재산상의 피해 앗차사고 부상이나 피해가 없는 사고 This study is based on these theorems. H.T. shows that there is a strong relationship among classified accidents. Say, the number of major accident, light accidents and near misses is 1 to 29 to 300. Similarly in Bird’s theorem, there is a strong relationship among serious or disabling, minor injuries, property damage and incidents with no injury or damage. Like here. These theories imply that hidden accidents cause real accidents. In other words, if we can reduce near misses, then we can inherently reduce major accidents.

10 2. 이론 1)사고유발 요인 모델 (Groeneweg) <안전 정보 관리 시스템>
안전 관리자(위험 판정) 위험요인 (G.F.T.) 파악 조직안전상태 분 석 규정을 벗어난 행동에 대한 심리적 징후 파악 심리학적 분석 불안전한 행동 분석 규정에 벗어난 행동 파악 운전 이탈 파악 감사 Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident. 방어벽 사고 재해 사고 분석 <안전 정보 관리 시스템>

11 3. 앗차사고 관리 필요성 1) 앗차사고의 분석 실제사고 하나는 많은 앗차사고와 연계 미국원자력 운전협회 :
원자력사고와 앗차사고를 비교한 결과 원인요소및 상황요소에 특별한 차이점 없음 앗차사고의 원인을 찾아 제거  실제사고 감소 Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

12 3. 앗차사고 관리 필요성 2) 공정안전관리제도 12가지요소
Hazard analysis and assessment report Procedure and manual for safety operation Procedure and specification for mechanical integrity Safety control procedure for contractor’s work Procedure for management of change Process safety information Emergency planning and response Procedure for hot work permit Education and training plan Process incident investigation plan 앗 차 사 고 관 리 채 택 Before talking about this study, I’ll introduce you several definitions. We can simple classify safety incidents by the following three: NM, LA and MA. The definition of Nmes is … Light accident is defined… Major accident is defined… This is a strong relationship among these three and using this relationship, we invented a new way of reducing sagety accident.

13 4. A기업의 앗차사고 분석사례 1) 실제 현장의 데이터로 증명 중대재해 : 경재해 : 앗차사고 결과는 하인리히의
( 6개월 동안 ) 중대재해 : 경재해 : 앗차사고 : : : : 결과는 하인리히의 법칙에 근사한 1 : 29 : 300 2) 사고의 분류 사고유형별 : 추락, 충돌, 붕괴, 화재, 기타 직접원인별 : 불안전 상태, 불안전 행동 간접원인별 : 기술, 교육, 관리 기인물별 : 공정 시설, 위험물 등급별 : A ~ E 등급으로 분류 In order to prove H.T., we collected industrial data from a petrochemical plant, And the total number of data for 6 months is 22 hundreds. And the ratio of major, light and near misses is 6 to 180 to This is 1 to 30 to 339. This is similar to H.T. Here we confirmed that there is a strong relationship among classified accidents. So we classified accidents in detail, like here. Accidents are classified by their results such as fall,.. and many others. Also the same data are classified by direct cause as unsafe.. And they are classified by indirect causes such as.. Also primary causes by process facility, Hazards are classified from A to E.

14 4. A기업의 앗차사고 분석사례 3) 사고결과에 따른 분류 사고발생 건수 결과

15 4. A기업의 앗차사고 분석사례 4) 사고의 직접원인에 따른 분류 사고발생 건수 불안전 상태 불안전 행동

16 4. A기업의 앗차사고 분석사례 5) 사고의 간접원인에 따른 분류 기술적 원인 교육적 원인 사고발생 건수 관리적 원인

17 5. 앗차사고 관리 시스템 1) Hazard Analysis Management System의 필요성
수천 ~ 수만 건 / 년 단위공정 수 많은 안전관련 데이터의 관리가 매우 어려움 수 많은 데이터를 저장, 분류, 분석, 관리할 프로그램이 필요 잠재위험의 파악 필요 투자 우선순위 제안 안전성 향상의 정도를 쉽게 파악(가시화) NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

18 5. 앗차사고 관리 시스템 2) HAM시스템의 구조 So far, I’ve explained the definition of NM and the industrial correlation. Now, I will show you the s/w that we implemented to manage NMes. The name of this program is yHAMS (yonsei…). This system includes two main modules, RDS(Relative database supervisor) and CSS(calamity superintendent sys.). You can regard RDS. as data storage. It classifies data as 7 states. As ... CSS is a system manager that receives accident data and shows the result of data analysis as a user requested. So, the results are...

19 5. 앗차사고 관리 시스템 3) 결과 보고서 화면 예시 (Explosion) (Others) (Stenosis) (drop) (Hazard Material Contact) (Poisoning, Suffocation) (Fall) This is another report with graph and tables.

20 5. 앗차사고 관리 시스템 4) 결과 그래프 예시 Tank (Including Storage Tank) (drop) (Hazard Material Contact) (Stenosis) (Poisoning, Suffocation) (Explosion) (Others) This is another computation of the classification. (Fall)

21 5. 앗차사고 관리 시스템 5) 기인물 및 등급에 따른 사고발생 건수 예시

22 6. 사고유발요인 모델의 필요성 전체 재해의 90% : 작업자의 불안전한 행동 인간 실수 분석 : 경영 시스템상의 문제
인 간 실 수 : 앗차사고로 연결 사고 유발요인 모델 활용  경영시스템상의 문제 파악  대책수립 앗차사고 감소 NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

23 7. 사고유발요인 관리 시스템 1) 위험요인 분야의 정의
NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

24 7. 사고유발요인 모델 2) 적용 결과 분석 조사대상 : 조사인원 62명, 답변자 55명(회수율 88%) A공장 9개 부서
조사대상 : 조사인원 62명, 답변자 55명(회수율 88%) A공장 9개 부서 조사방법 : 설문조사 분석기간 : ~ 분석항목수 : 12,100 문항 (55명 11요인 20문항/요인) 분석프로그램 : 엑셀 위험요인별, 사업장별 수준 부서별, 개인별 수준 NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

25 7. 사고유발요인 모델 3) 분석결과 예시 (A회사 전체)
NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

26 7. 사고유발요인 모델 4) 분석 결과 예시 (A회사 전체)
NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

27 7. 사고유발요인 모델 5) 부서별 위험요인별 평균수준(SM)
NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

28 7. 사고유발요인 모델 6) 응답자 개인별 위험요인별 수준(SM)
NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

29 7. 사고유발요인 모델 7) 활용 방안 사업장별, 부서별, 개인별, 위험요인별 수준 평가 재해예방 자료로 활용
동종재해 예방대책 수립시 적용 NM management program is needed for handling tens of thousands accident records. Without the program, it is difficult or almost impossible to manage those data. The program must be able to save, search, .. data. As a result, it should identify… and also Suggest priority… So we can easily analyze the.. improvement. As you already know, it is very difficult to manage numerous accident data, since chemical plants include a lot of equipment and the accident data are accumulated as time goes by. So it is almost impossible to handle those many data manually. This is the reason why we developed the s/w.

30 8. 결론 앗차사고와 실제사고 사이의 관계 증명 사고예방 측면에서 투자 우선순위 선정 및 투자가 필요한 단위공정 제시
새로운 위험관리기법의 제시 사고유발요인모델 활용, 경영시스템상의 문제점 파악 체계적인 분석, 대책수립, 앗차사고를 포함한 효과적인 재해예방 가능 The industrial data confirms that there is a strong… So we developed a new program to handle numerous data efficiently. With the program, we are able to suggest the priority financial,,, Currently we are working on analyzing the relationship between the investment cost and the reduction of safety accidents. This program can also provide history data for,, Just for information, I set-up a new business and sell this program to industries. Thank you, for your attention.