High Reliability Process Industries: Individual, Micro, and Macro Organizational Influences on Safety Performance
Authors：David A. Hofmann, Rick Jacobs, and Frank Landy
Source：Joumal of Safety Research, Vol. 26, No. 3, pp. 131-149.1995
這篇是15年前的文獻回顧彙整型的文章
來看看15年前的知識水準（social-organizational factors on safety），另外需特別注意這篇文章是針對化工/核能產業（能否推論到整個製造業，可能存有爭議）
Abstract：
Within the process industries (e.g., chemical and nuclear power plants), safety is of paramount importance. Although there has been much research investigating safety issues in these industries, the current paper reviews and integrates literature pertaining to individual, micro organizational, and macro organizational influences on safety. This paper can serve as a starting point for continued consideration of the influences of social-organizational factors on safety.
SAFETY CONSIDERED FROM THE INDIVIDUAL LEVEL
被濫用的人為疏失（human error）一詞
(the term “human error” is frequently found in “root cause” statements).不知ㄧ票人是否都被杜邦的分類（90%的人為疏失、10%設備故障=千錯萬錯都是人的錯）所誤導
For example, Kletz (1985) reports that accident statistics, mostly from the chemical and oil industries, indicated that 50% to 90% of industrial accidents were due to “human failing.” As we will see below, these “human errors,” while often being classified as the root cause of accidents/failures, may be influenced by larger, orgauizational issues, some of which can actually implicitly encourage unsafe acts.
也間接導致安全文化/氣候學派的興起
Cognitive Ermrs and Schemata
文中舉的例子一個化工廠員工，無意識地依平時程序打開一間未洩壓房間的門，結果當然發生意外，像這種無意識的疏忽，即便歸為人為疏忽也無法防止意外發生
From Kletz’s (1985) perspective, this chemical industry accident was inevitable because human errors, or slips, will occur He stated, “we now see that. . . it is inevitable, sooner or later, that an operator will forget that he has not opened the vent valve and will try to open the filter while it is still under pressure (p. 7).
Although committing procedures to written form makes the sequence of operations clear, the problem facing those concerned with safety is that operators follow the steps contained in the written procedures, not necessarily that they remember the appropriate sequence of steps.（雖然安全程序都有寫，但難以確保員工遵守或記得）
Both Kletz (1985) and Reason (1990) conclude that slips will occur but suggest that these slips can be minimized through better training and instructions or procedures. This brings up a second, critical issue.=>如何判定當事人的資格與能耐competence?
The individual, cognitive, and knowledge- based factors are critical factors influencing safety performance in process industries.
聯想到國內法規寫的承攬商危害告知，看來還是只管完成形式上的上課和測驗，難以管制承攬商作業人員的認知和能力是否足夠…
Safety motivation and attitudes
An additional issue that must be considered at the individual level of analysis is the pattern of attitudes of the employees as it applies to health and safety.
The motivation to perform a job in a safe manner is a function of both the individual’s own concern with safety as well as management’s expressed concern for safety.
造成安全態度不佳的因素
Wright (1986), through his detailed analysis of the accident reports in the oil industry, discovered two underlying themes in all of the accidents: (a) an emphasis on “speeding-up” the work processes, and (b) failed communications.
可以說是求快和cost down+老人欺負新人/溝通不良？
當然這其中很關鍵的是基層主管的認知
Perhaps the best indicator that faulty communications plays an important role in accidents is the fact that the majority of the victims in the reports reviewed by Wright were contractors.
不意外的受害者當然是承攬商
Thus, the situations that produced these accidents were both normal and abnormal. They were normal in the sense that all the situations producing the accidental deaths were commonly occurring work situations. They were abnormal in the sense that the outcomes (i.e., the deaths) were unfortunate, unintended, and unforeseen.
In other words, unsafe acts were built into the daily work routines and it is precisely because these acts became “routine” that potentially disastrous outcomes were not anticipated.
造成人為疏失的來源：不幸的巧合、無意識的行為舉止、不可預見的狀況
No individual ever intends to have an accident.
Individuals have intentions to behave in certain ways and accidents can be the unexpected outcome of an individual’s behavior. In other words, the meanings of actions (e.g., unsafe behaviors) must be viewed in light of the individual’s motives, purposes, principles, beliefs, and attitudes (see Taylor
事出必有因，當事人之所以會有不安全的行為，背後一定有其邏輯思維和價值判斷（EX：求快、貪圖方便…）。
這又引申到低階主管和現場人員的關係與默契
Two important conclusions can be inferred from the intervention regarding successful programs. First, workers must see managers devoting time and attention to safety issues, and second, workers must be permitted to help shape interventions rather than simply playing the more passive role as ‘recipient” of the intervention.
(d) the desire to make the job a little easier.
看這篇回顧，也才更了解一些安全文化/氣候構面的來龍去脈和其血肉
也難怪談安全氣候，就是心理學和行為學派的學者猖獗的時候
SAFETY CONSIDERED FROM THE MICRO ORCANl2IATlONAL LEVEL
這個部份的影響因子含：
高階主管的重視、工會和勞資間的溝通、事業單位的自主管理、安衛政策和員工參與
當然造就以上事業單位安衛措施的背後，其實是政府的法規和行政措施
相關的公司個案研究，延伸出兩大主題：相關技術和如何改善員工的態度知識
First, a general lack of respect for the technology emerged as one theme.
The second general theme was employee attitudes and/or knowledges.
Other conclusions regarding health and safety
policies reported by Dawson et al. (1988) were:
l There was a considerable feeling that not enough was done on the shop floor to train and refresh workers and supervisors about how to secure good standards of health and safety, particularly following the introduction of new materials or processes
l Expressions of satisfaction with the organization and arrangements for safety and health reflected low or vaguely conceived expectations and a dearth of standards against which present performance could be evaluated.
l Generally older workers, who were also higher in the organizational hierarchy, were more satisfied with the health and safety arrangements than younger workers or individuals in lower levels of the organization.
l In two of the organizations, there was a general feeling that no one of any importance really seemed to care about health and safety and that little was done to commit resources to secure improved standards
這個部份就比較有時代上的落差，經過這些年BS-8800和OHSAS-180001都已經算是成熟的系統
Finally, it should be mentioned that although the plants in the Dawson
這裡點出了賺不賺錢和事業安全績效間的關連性
If money cannot be spent on improved equipment one might argue that this is all the more reason why people should be ‘more careful,’ and ‘more safety conscious,’ in order to compensate, through behavior, for physical or technical shortcomings.
In fact, the opposite seems to occur.
When suggested actions are not taken to make plants physically more safe, people too appear to become less concerned with observing safety procedures. If less is being spent on safety it appears to become a less important issue in the minds of all concerned. (p. 88)
當然願意在工安上投資，也才是管理階層真正的承諾(而非嘴砲)，而在不景氣的時候更能看出管理階層承諾的真心
而本著後見之明來看以上因果關聯猜測，會覺得過分簡化。（因為後來的實證研究已經證實：景氣越好、工安事故率越高=>換言之，肯不肯花錢投資工安和事故率之間的關連性，沒有景氣因素來的大）
另外作者認為安衛績效的真正核心還是在於勞資雙方的互動，而非政府的干預與管制（芝加哥學派的想法）
Although the role of governmental agencies is no doubt important in “encouraging” self regulatory activities, the critical linkage, with respect to health and safety, is still between management and labor. The behavior of management regarding health and safety issues can send many messages to workers.
SAFETY CONSIDERED FROM THE MACRO ORGANl2IATlONAL LEVEL
We have proposed that there are both individual (e.g., attitudes and training) and micro organizational (e.g., management support and safety committees) influences on the effectiveness and safety performance in an organization.
The following review briefly outlines the typical organizational structure of high reliability organizations, profiles the structure of a single high reliability organization, and then moves into a discussion of the effects on organizational culture on organizational reliability.
作者這裡指的Macro 是指組織結構和文化=>而非產業文化、民族性、國家政治法律規範等(換言之，還是屬於個體的層次)；這樣的分類，現在看來有些怪，但還是很欽佩這些前輩大師對於組織的內部運作能夠有這些洞見（而這些洞見源自於組織理論）
Macro Organizational Level是指員工的專業分化程度、作業程序間的關連性、決策機制、水平和從垂直的溝通/訊息傳遞等 之間的整合和衝突
From the work of Permw (1984). as well as organizational theorists (Scott, 1987), one can predict that the structure of high reliability organizations will be characterized by a high degree of specification, interdependence, centralization, as well as an emphasis on expending resources to ensure the coordination of the activities of different departments.
產生的一些有趣衝突和對立（paradox）
中央集權VS組織專業分工（該聽行政大老闆還是總工程師？）
Increasing the redundancy in a complex system may, beyond a certain point, actually decrease its reliability.（提高系統備份單元，反而降低整體可靠性）
容錯彈性VS精確控制（當面臨環境的不確定因素時）
The third paradox discussed by Roberts and Gargano (1990) was that organizational theory would dictate that uncertain environments should result in rat& loosely coupled organizations. This loose coupling enables the organization to adapt to the environment more quickly. High reliability organizations, while typically operating in an uncertainty environment, ate rather tightly coupled.
分工VS專業（只有高中畢業的工人如何能夠勝任/理解複雜的製程狀況）
Roberts and Gargano rhetorically ask if these operators, who may only have a high school education, can handle the complex, unforeseen, interactive, and unpredictable events that occur in nuclear power plants.
發揮最大產能還是保持彈性餘裕？
The paradox is that most high reliability organizations operate at near capacity thereby reducing their flexibility.
許多公司都行相關的SOP和程序書，然而這些SOP卻無法讓人知道整體的運作方式（檯面下的潛規則和人跟人之間的交情默契）
Standard Operating Procedures (SOPS), which describe the process of integration, never explained how the entire system worked together to operate smoothly
Culture as a Source of High Reliability
Weick (1987) discussed the concept of high reliability as it is related to, or subsumed under, organizational culture.
現在個人感覺其實用所謂的文化來涵蓋解釋安全績效，其實跟文章一開始對於所謂”人為疏失”的批判一樣；感覺都只是用個「詞」來涵蓋唬爛，與其用文化這個詞來涵蓋加總，還不如說是組織裡所有人員對於安全的態度和價值觀的分寸拿捏，決定的公司總體的行為模式，而行為模式決定了意外事故發生的或然與必然。
（用以上說法來看現行安全文化與氣候的研究，或許最大的罩門在於內部效度與抽樣結果的代表性：某家公司單一工安人員填寫的問卷結果，如何能夠代表該公司其他所有人對於安全的價值觀和態度？乃至於單一時點的問卷結果，如何反應個人或群體在觀念和態度上的動態變化？）
另外這個部份作者相關研究引用的例子是美國海軍，當然美國海軍這個行業與職務，具備一些非常鮮明的組織特色與文化，但相較於一些製造業公司而言，或許不存在”顯著”的「文化」可供探討；也或許不同世代之間的探討比較才能用「文化」這個詞。（EX：化工廠，四年級員工和七年間員工的安全文化認知差異比較）
The Influence of Regulation on Structure and Climate
略
CONCLUSIONS AND IMPLICATIONS
Individual
l Employee attitudes: Several safety related attitudes have been linked to the (un)safe operations of process industries.
l Employee behaviors: Two very specific patterns of behavior emerged as potential safety related hazards. From our review of the process industry literature it appears that there are two related types of problems that emerge from employee familiarity. The first deals with routinely using short-cut methods (methods that skip steps or fail to provide adequate safety margins) and adopting these as unofficial, standard operating procedures.
l Employee knowledge: This category represents a broad range of threats to safe operations related to employee knowledge and/or information.
Micro organizational
l Self-regulation: Our review of the literature pointed directly to the need for an organization to take responsibility for health and safety activities and to make it a “within the organization” activity. It was clear that an organization suffers when an external agency is seen as the underlying rationale for operating in a safe manner. One method for insuring the self-regulation perspective is the creation of management and labor teams that mutually agree on safety procedures.
l Organizational policies: Related to the self regulation dimension, our review uncovered the need to have workers view safety policies as those created by and in the best interest of the organization. This principle applies to policies that are generated by the organization or by an external agency. The key variable is the degree to which the organization can internalize the policy by seeing the link between compliance and safe operations.
l Design of work environment: Kletz (1985) demonstrated that redesigning the work environment could reduce or prevent human errors; “change the situation not the person.” For some problems it is possible to simply redesign the work environment. The point here is that as we seek to better understand the less obvious organizational factors that influence safety, we should not overlook the more straightforward human factors changes that can have a profound impact on safety.
l Safety representatives: An organization can improve its safety attitudes and behaviors by appointing safety representatives. Management attitudes: Once again, it becomes possible to develop a list of potential contributors to the concept of “safety culture.” This time the entries come from management activities and can be seen as influencing the individual attitudes listed earlier
l Accountability: One important feature of a safe organization is its vigilance in maintaining records of safety related problems/ achievements. The need for accountability at the governmental, organizational, and unit management level regarding health and safety issues points to the importance of having records that can be used to better understand safety related performance.
MACRO ORGANIZATIONAL
l Technological complexity: There is no dispute that process industry facilities represent highly complex technological systems.
l Work force specialization: Within technologically sophisticated organizations, there is a highly specialized work force with very departmentalized responsibilities. This serves to be a positive impact on safety under normal operating conditions when the resident expertise can be applied to the system where it is needed. Engineering has its expert engineers, maintenance has its experienced and knowledgeable crafts personnel, and operations has its well trained crews.
l Centralization & De-centralization: In most high reliability environments them exists a simultaneous push for de-centralization in a structure that demands centralization. Based on this review, it appears that an organization is most successful in dealing with this issue if it is able to establish an accepted process of centralization. Once this has occurred the organization is in a better position to move into decentralized decision making when it is required.
l Vertical and horizontal communication: Given technology, specialization, and tightly coupled nature of the high reliability organization, vertical and horizontal communication is critical for success.
l lightly coupled: Very little slack in the system; occurrences in one system influence other systems. With this “dependency,” the obvious problems of correlated errors, domino effects and interrelated mishaps become paramount.
l Redundancy: Redundancy of both technology and management/decision making structures is a key descriptor of any high reliability process industry. The idea is to multiply the lines of defense in an effort to reduce the likelihood of a problem.
-----------------------------------------------------------
總結感想：
- 讀的很累但頗有收穫的文章，這種涵蓋員工個人、事業單位個體到組織結構的研究果然很有挑戰性（較近期的研究更納入政治經濟總體層面因素的探討）
- 學術研究果然是站在巨人肩膀上，感謝這些未曾沒面前輩的洞見，而環安也的確是一綜合應用學門，借用了更多來自心理學、行為科學和組織理論大師的經驗智慧，在15年後的現在看來，許多想法和建議還是很受用。
- 想到一位友廠同業很棒的比喻：工廠製造生產和工安之間，有如賽車一般=>目標是在不出事的狀況下，用最快的速度達成目標；所以過分強調安全的賽車手雖然可以活命、但不可能享受冠軍的榮耀，而一昧求快的車手，不但拿不到冠軍，還可能會要做頭七。
- 用以上角度來看，其實工廠出事就跟賽車場上發生撞車的車禍一樣、一點都不讓人意外；然而此一角度卻也預言了=>撞車事故(工安事故）一定會發生，難以防範，所以某種程度來說，工安管理其實是徒勞無功的？！.
- 擔任環安主管真的是一門學問與藝術=>一門知道如何拿捏分寸（偷工減料）的學問與知道什麼時候該提醒老闆當膽小鬼的藝術…
沒有留言:
張貼留言