From accountability investigations to problem investigations —— Reflections and recommendations based on systems theory and system safety theory
Thread Content
Part 1: Investigating and handling production safety accidents is an important aspect of work related to safety in production. Innovations in accident investigation policies and methods are of great significance for improving the institutional framework for safety in production in our country. In light of the overall situation of work safety in China and the current state of accident investigation and handling, this paper analyzes the trends and causes of particularly serious accidents in China over the past three years, and summarizes the basic patterns and common issues associated with such accidents. Based on the concept of system security, the author believes that security is an inherent attribute of the entire system, and system vulnerability is the main cause of accidents; therefore, the occurrence of accidents should not be simply attributed to the fault of certain individuals or specific events. Since the establishment of China’s work safety supervision system in 2001, the work safety situation has remained stable and continuously improved. However, the risk of major and extremely serious accidents remains high, and the overall work safety situation is still grave. Looking back at the historical development of work safety in our country, there are many experiences and lessons worthy of careful summarization and assimilation, so as to further improve and innovate our work. I. The severe situation regarding work safety represents a major challenge to China’s economic and social development. The frequency of extremely serious accidents is on the decline, but there is significant variation in the data; currently, the risk of such extremely serious accidents in China remains high. Although the particularly serious accidents in recent years differ greatly in terms of time, location, industry, immediate causes, and disaster patterns, they share certain common regularities. This series of similar problems reveals widespread systemic vulnerabilities in work safety. We should focus our efforts on the overall properties of the system, working to improve the level of risk management for major accidents within the entire system as well as its resilience in dealing with such accidents. Since the reform and opening up, the frequent occurrence of major and severe production safety accidents has been a persistent problem affecting China’s economic and social development. At the beginning of this century, our country established a work safety supervision system that remains in use to this day. The Party and **took a series of significant measures to strengthen work safety efforts, achieving notable results; the most prominent indication of this is the continuous decline in the total number of work safety accidents (see figure below). From 2005 to 2015, the number of deaths resulting from work safety accidents in China continued to decline; the death toll was 127,089 in 2005, while it dropped to 66,182 by 2015. However, as can also be seen from the figure below, although the frequency of major and extremely serious accidents is also on a downward trend, the data exhibits significant fluctuations. The coefficient of variation regarding the annual frequency of extremely serious accidents reaches 0.5. This indicates, on one hand, that such accidents occur in a discrete and random manner with low probability; on the other hand, it also suggests that the risk of extremely serious accidents in China remains quite high at present. Preventing and curbing major and extremely serious accidents has become a top priority in China’s work related to work safety. Around 2005, there were about 10 such major accidents per year in China; although this figure has dropped to around 5 per year in the past five years, from 2013 to 2015, there was one major accident with over 100 deaths each year, incidents of particularly severe nature with extremely negative consequences. Preventing and curbing major and severe accidents has become a top priority in China’s work on work safety at present. In fact, the occurrence of particularly serious accidents is not entirely without patterns, but the problem now is not a complete lack of knowledge about their causes; rather, many of the understandings are misleading. Based on a systems-thinking approach involving case analysis and comparative evaluation, the author analyzed past major accidents. This may offer new insights into the risk management of extremely serious accidents. The author made a brief analysis of 12 extremely serious accidents that occurred from 2013 to 2015. Of these 12 particularly serious accidents, only 4 are typical production safety accidents, while the other 8 have public safety characteristics to a certain extent. These 12 particularly serious accidents resulted in a total of 860 deaths and 1,218 injuries. Following these 12 particularly serious accidents, a total of 845 people were held accountable. Furthermore, although the investigation and handling reports for these 12 particularly serious accidents all identified the distinct direct causes of each incident, without exception they were all classified as liability accidents, which naturally led to serious accountability actions being taken regarding these particularly serious incidents. A series of similar problems reveal systemic vulnerabilities that are widespread in work related to safe production. Accidents in modern society are generally highly complex; catastrophic behaviors and vulnerabilities arise in almost all fields, from nature to society, and they intersect and merge with one another. Systemic vulnerabilities exist at all levels of safety management, as well as throughout the entire process of emergency preparedness, response, and recovery in the face of sudden incidents. System vulnerability is an objective condition that exists prior to an accident, and it only \"emerges\" during the course of the accident. From a systems theory perspective, the emergence of system vulnerability can exhibit different behaviors depending on the system’s characteristics and changes. Before an accident occurs, system vulnerabilities remain latent; this uncertainty significantly increases the “vulnerability” to accidents. Such emergence in the system depends not only on whether risks are perceived or identified before an accident occurs, but also, and importantly, on whether appropriate control measures are taken once they have been identified. Only by identifying, assessing, and controlling the potential system vulnerabilities that lead to accidents before they occur can the resilience of the entire society in dealing with accident risks be comprehensively enhanced. Over the years, whenever a major accident occurred, relevant Chinese authorities conducted serious and thorough investigations and handling. They drew lessons from each incident, earnestly summarized the painful lessons learned, took corrective measures, strengthened a range of preventive controls, and imposed ever-stricter accountability. These measures have indeed helped to advance the work on ensuring safe production, but the risk of severe accidents remains uncontrolled. From 2013 to 2015, the 12 particularly serious accidents that occurred in China may seem quite different in terms of when and where they happened, the industries involved, their immediate causes, and the nature of the disasters. However, upon closer analysis, it becomes apparent that there are certain common patterns underlying the causes and circumstances of these disasters: failure on the part of enterprises to fulfill their responsibilities regarding safe production, insufficient attention paid to safety production at the local level, inadequate supervision of safety practices, poor handling of potential major accident risks, ineffective emergency response measures, as well as weak infrastructure and low safety awareness among relevant personnel. This series of similar problems reveals widespread systemic vulnerabilities in work safety. System vulnerabilities are essentially structural risks of a system. System vulnerability is an emergent property of the system as a whole, rather than a characteristic of local areas or individual units. Therefore, in both pre-accident risk management and post-accident investigation and handling, attention should be focused on the overall characteristics of the system, with efforts being made to improve the level of risk management for major accidents within the entire system as well as its resilience in dealing with disaster situations. II. Production safety is a systemic attribute. In the context of today’s highly complex economic and social development, traditional theories and methods of production safety management are gradually becoming ill-suited to the demands of the new situation. Modern system security theory holds that security is an attribute of a system. At present, the problems revealed by the grave situation regarding work safety in our country are by no means merely management issues within individual entities; at their root, they stem from widespread weaknesses in the foundation, imperfect institutional mechanisms, and a lack of rule of law development. The significance of risk governance and system security lies in identifying and controlling risks that could lead to serious accidents before they occur, thereby preventing such accidents from happening. Experience tells us that for a recurring phenomenon, we should seek explanations from patterns, while for widespread problems, we need to analyze the causes from the systemic structure, including institutions and mechanisms. Currently, China’s work safety efforts are facing unprecedented pressures and challenges. The continuous advancement of reforms and the accelerating pace of urbanization, the emergence of new modes of production, as well as the profound context of global economic integration, have given rise to some highly complex new issues in work safety. As a result, traditional theories and methods of work safety management are increasingly finding it difficult to meet the demands of the new situation. The author believes that the Heinrich accident chain model, which has long been regarded as a classic, Bird’s theory of management decision-making, the Swiss cheese model of accident causes proposed by Rien, as well as theories regarding causal relationships and accident chains that have been in use for many years, are no longer fully suitable for preventing major accidents, responding to them, and conducting investigations into such accidents in the highly complex context of current economic and social development. Safety is a property of the system, not of individual elements or parts. Modern systems safety theory holds that safety is a property of the system, rather than of individual components or local areas; both safety and accident risk are emergent properties of the system. The analysis of several major accidents that have occurred in recent years shows that the occurrence and development of such accidents involve complex economic, social, cultural, and technological processes. It is necessary to focus on analyzing how the overall operation of the system leads to accidents and losses, rather than simply attributing the causes of these accidents to certain individuals or factors. Most major and severe accidents occur as a result of nonlinear interactions and multiple couplings within complex systems. Simple accident chain models and causal relationship determination methods suggest that an accident is the result of multiple individual events occurring in sequence over time; therefore, it is believed that breaking the failure chain can prevent accidents from occurring. Modern system safety theory holds that accidents arise from the interactions between various components of a system; they generally cannot be attributed to a single variable or individual local failures, but rather there is almost certainly some structural risk (systemic risk) present. A major accident is less the simultaneous occurrence of multiple independent events and more a transition of the system toward a state of increased risk over time. Complex systems inherently exhibit a tendency to gradually shift toward risk, and this tendency is detectable and predictable. It can be controlled through appropriate system design and pre-established risk detection indicators, thereby preventing accidents from occurring. Behind every accident, there are almost always very complex reasons. On the surface, factors related to individual or departmental responsibility seem to play a significant role. This is why, after any major accident occurs, it is almost invariably referred to as a “negligence-related accident.” However, this uniform perception may be overly one-sided and simplistic. From the perspective of system safety theory, in the complex progression of major accidents, little emphasis is placed on the responsibilities and obligations of individuals, unless such individuals have a deliberate motive to cause the accident, or are suspected of committing offenses such as negligence and dereliction of duty, which play an irreplaceable role in the occurrence and development of the accident. In fact, relying solely on officials’ dedication to their duties cannot completely prevent accidents from occurring. For example, under significant commercial pressure, the risk of production safety accidents in enterprises increases markedly, and it is difficult for individual efforts to change this situation. Risk governance and system security cannot be achieved merely by complying with laws and regulations. Laws and regulations are merely baseline standards for safety, not high-level requirements to prevent accidents; there is significant room for improvement between these baselines and higher standards. Numerous accident investigation results have repeatedly shown that safety management based solely on compliance with laws and regulations is far from sufficient for highly complex systems. This is because such compliance cannot address the high levels of uncertainty arising from the complexity and emergent properties of these systems, nor can it detail specific operational procedures. Moreover, new hazards that arise from the introduction of new technologies and situations cannot be covered by existing laws, regulations, and standards. Major accidents occur as a result of the failure of system constraints or errors in feedback. As a complex and open macro-system, safety management in such systems emphasizes not only integrity (where the whole is not merely the sum of its parts) but also the hierarchical structure of the system; hierarchy is a fundamental characteristic of systems theory, systems engineering, and system safety. According to the theory of system security, a security management system should be divided into several structural or organizational levels, with security being achieved through the constraints imposed by higher levels on lower levels. Regardless of the hierarchical structure, it is through layer-by-layer constraints from top to bottom that a three-dimensional, comprehensive security management system network is formed. The role of the lower layer on the upper layer is not to impose constraints, but rather to provide the enforcement for those constraints as well as a feedback mechanism – that is, to reflect issues promptly and accurately and ensure effective information exchange. The lower layer also does not impose any constraints on the system; security emerges in the system when the constraints imposed by the upper layer on the entire lower layer are stable and effective. When the constraints exerted by the upper layer on the lower layer fail, or when the feedback from the lower layer to the upper layer is incorrect, this can lead to an increase in the risk of accidents within the system – what is known as the emergence of risks within the system. In severe cases, it may even result in the failure of the overall system’s constraints, leading to major accidents. From the perspective of system safety theory, major accidents occur due to the failure of system constraints or feedback errors; this dynamic process depends not only on a single variable or merely local factors. Even if a subsystem or a particular part of the whole is optimized, it will not result in an optimal system. In reality, it’s difficult to completely avoid individual human errors and localized failures; however, redundancy in the system architecture and fault-tolerance mechanisms, along with the system environment, can significantly reduce the risk of accidents. For example, through an in-depth analysis of the particularly serious accident of the leakage and explosion of the Sinopec Donghuang oil pipeline in Qingdao, Shandong on November 22, it was found that the source of the hazard was the flow of over 3,000 tons of light crude oil from the leaking Sinopec pipeline into the underground drainage channels in the industrial zone. As the petroleum gas evaporated, a large amount of explosive gas was formed in those enclosed underground spaces. No effective risk control measures were taken to address the explosion hazard after the leakage occurred, and it was not until 10 hours later that a major explosion took place. Other deeper underlying causes of the accident include the lack of long-term planning for the development areas and crude oil storage and transportation, as well as the outdated condition of the oil pipelines that have been in use for over 20 years ； There was no strict safety regulation for major hazard sources; emergency responses were not initiated promptly in the event of a leak, nor were any measures taken immediately to protect the public. In summary, the technical causes of the accident were inadequate risk management and insufficient emergency preparedness; from the perspective of system safety theory, it can be determined that the risks inherent in the system structure gradually evolved into a severe accident. Rather than attributing the essence of the accident to the negligence and dereliction of duty on the part of grassroots officials and enterprise staff, it is more appropriate to view it as a systemic vulnerability that emerged at the relevant levels as a whole. In fact, it is difficult to confirm which official or which action involves clear illegal conduct, let alone establish a direct and inevitable causal link between such individual actions and the occurrence of the accident. More emphasis should be placed on identifying one’s own shortcomings, as this is crucial for preventing similar accidents from occurring again. Years of experience in safety production have shown that relying solely on holding individuals accountable may temporarily ease societal pressure and address some superficial or partial issues, but it fails to delve into the root causes of accidents, nor can it resolve the deep-seated problems related to systematic, fundamental, and institutional aspects. The likelihood of major accidents occurring due to individual or single factors is very low, and the driving forces in such cases are also limited. It is the systematic structural flaws stemming from organizational, social, and cultural factors that are the main reasons for the continuous occurrence of accidents, and even for the repeated occurrence of similar accidents. At present, the problems revealed by the grave situation regarding work safety in our country are by no means merely management issues within individual entities; at their root, they stem from widespread weaknesses in the foundation, imperfect institutional mechanisms, and a lack of rule of law development. This is of overall significance and widespread applicability in the field of work safety in our country. Historical experience has shown that after major accidents, if the causes of those accidents are attributed solely to \"fault,\" even if the allocation of responsibility is fairly even and the level of accountability is moderate, at best this merely leads to an awareness of the seriousness of the problem, with only lessons learned and some of the causes of the accident identified. There is an important principle in system safety theory: if the risks associated with a system are not properly recognized, or if no control measures appropriate to those identified risks are put in place, it is essentially equivalent to \"assuming nothing or waiting for an accident to occur.\" From the perspective of risk governance, any production safety accident is knowable and preventable. Regarding highly uncertain accidents that occur in complex systems, it is indeed difficult to make accurate and specific predictions in terms of time and space beforehand. Moreover, the more complex a system is, the more prone it is to system failures due to its inherent vulnerabilities, ultimately leading to major or catastrophic accidents. Relatively speaking, systems with high complexity and tight coupling have a higher risk of experiencing failures. Complex systems can introduce greater uncertainty, and nonlinear interactions occur when multiple factors are closely interlinked, thereby increasing the risk of accidents. As the renowned security expert James Lesson once pointed out, “Complex and tightly coupled systems can suddenly give rise to serious accidents.” On June 1, 2015, the passenger ship Dongfang Star sank in the Jianli section of the Yangtze River in Hubei, resulting in 442 deaths. This disaster is a typical example of such a “sudden occurrence of a severe accident.” The essence of this incident should not be attributed solely to the height of the waves or the strength of the wind, but rather primarily to the lack of reliability in the inherent safety performance of the ship’s hull. There are safety defects in the design and modification of the hull, as well as obvious gaps and risks in the navigation safety management system. Passenger ship managers lack risk awareness and emergency preparedness; there are also significant problems in responding to sudden dangers. In cases like this, we should pay more attention to identifying our own problems, especially focusing on the system vulnerabilities that arise during an incident; this is crucial for preventing similar accidents from happening again. The significance of risk governance and system security lies in identifying and controlling risks that could lead to serious accidents before they occur, thereby preventing such accidents from happening.
Part 2: Our country has always imposed strict accountability on those directly responsible for serious production safety accidents. The accountability for work-related accidents plays a crucial role in protecting the safety and health of the general public, upholding the seriousness of laws and disciplines, enhancing the sense of responsibility among officials at all levels, and preventing the frequent occurrence of major and extremely serious work-related accidents. However, through an in-depth analysis of the investigation and handling processes of numerous major and serious accidents, the author found that if the intensity, scope, and methods of accountability enforcement are not properly managed, certain negative effects may ensue. The author believes that with the development of China’s economy and society, some provisions and requirements in State Council Order No. 493 have clearly become inapplicable under current circumstances; therefore, it is urgent to make certain revisions and innovations to the working system regarding the reporting, investigation, and handling of work-related accidents. Six suggestions for amending State Council Decree No. 493 ★ China’s safety supervision system should further clarify its role as the main body responsible for **administrative law enforcement**, gradually reduce its managerial responsibilities, and establish mechanisms to ensure that departments responsible for safety supervision have clear guidelines to follow. ★ When classifying accidents, economic losses should not be used as a criterion for determining the severity of work-related accidents. ★ A more scientific time limit should be set for accident investigations. ★ We should draw on foreign experiences where specialized accident investigation and analysis agencies are established by law; these agencies then organize relevant experts and representatives from various sectors to conduct independent accident investigations. ★ Some standardized and practically applicable procedures should be added. For example, provisions regarding how to manage media interviews and public opinion guidance can be added; in particular, regulations concerning the Internet and new media should be emphasized. ★ After an accident investigation, special attention must be paid to feeding back the lessons learned and applying them in practice as soon as possible, so as to prevent similar mistakes from occurring again. III. Accountability can be a double-edged sword for safety. Holding parties accountable for work safety accidents is undoubtedly an effective tool for promoting the rule of law in work safety. However, if the intensity, scope, and methods of such accountability are not properly managed, certain negative effects may arise. Excessive pursuit of accident liability invisibly increases the costs of safety supervision, hinders the determination of the truth behind accidents and a thorough investigation of their causes, and may mislead the public and the media. These issues not only dampen the enthusiasm of safety supervisors, but also hinder the identification of lessons learned from accidents, making it difficult to provide a basis for institutional improvements aimed at preventing future incidents. They are highly detrimental to promoting safe production practices throughout society as well as to the development of public safety. The rule of law is the most important and fundamental measure in work safety; therefore, punishing illegal acts and holding those responsible for serious accidents become important tools in ensuring safety through the rule of law, a principle that holds true worldwide. China has always imposed strict accountability on the direct responsible persons for major and extremely serious production safety accidents. As early as February 27, 1950, a gas explosion occurred at the Yiluo Coal Mine in Henan Province, resulting in 174 deaths. At that time, the Government Council took serious action regarding this accident; the responsible officials from Henan Province and related departments, as well as those involved in the coal mine, were held accountable under criminal and administrative laws. On November 25, 1979, the drilling rig Bohai No. 2 sank in the Bohai Bay, resulting in 72 deaths and significant property losses. The investigation concluded that the leaders of the Petroleum Industry Ministry bore significant and undeniable responsibility for this incident; relevant leaders of the State Council handled this serious matter inadequately. The then Minister of Petroleum Industry was removed from his position, while the Vice Premier of the State Council in charge of the petroleum industry was given a severe administrative reprimand for his role in the incident. On April 21, 2001, the \"Regulations of the State Council on Holding Administrative Accountability for Extraordinarily Serious Safety Accidents\" (State Council Order No. 302) was issued and implemented, further strengthening the accountability of leading officials. In April 2007, the State Council issued the \"Regulations on the Reporting, Investigation, and Handling of Production Safety Accidents\" (State Council Order No. 493, hereinafter referred to as Order No. 493), which established clear provisions regarding the procedures for reporting production safety accidents, the process of investigating and handling such accidents, the classification of production safety accidents, and the determination of liability for these accidents. These regulations formed the basic framework for the investigation and handling of production safety accidents to date, playing an important role in standardizing such efforts in China. Based on years of practice, the accountability for work safety accidents has played a very important role in protecting the safety and health of the general public, upholding the seriousness of laws and regulations, enhancing the sense of responsibility among officials at all levels, and preventing the frequent occurrence of major and extremely serious work safety accidents. It has thus become a powerful tool for promoting the rule of law in work safety. However, through an in-depth analysis of the investigation and handling processes of numerous major and serious accidents, the author found that if the intensity, scope, and methods of accountability enforcement are not properly managed, certain negative effects may ensue. This makes the pursuit of liability for accidents a double-edged sword. Excessive accountability for accidents inadvertently increases the costs of safety supervision. At present, a performance evaluation system for work related to safety production supervision and law enforcement has not yet been established in our country. Against the backdrop of a severe safety production situation, it is inevitable that there will be a high-pressure approach to holding those responsible for accidents accountable. Under such high-pressure and stringent conditions, the responsibility of supervisors for ensuring the safety of employees becomes greater, and the costs increase as well; yet the benefits do not increase in proportion, resulting in a severe imbalance between costs and benefits. No matter how hard or diligently cadres work, they won’t necessarily be praised or promoted in a timely manner if no accidents occur; but once an accident happens, regardless of how much effort they have put in, they will be held accountable (cost). This artificially created situation of coexistence of responsibilities and risks, along with the asymmetry between responsibilities and benefits, makes work related to safe production a high-risk endeavor. In such situations, some officials may, in order to protect their own interests or reduce the risk of damage to those interests, inevitably handle matters according to the principle of centralism and adopt harsh policies; as a result, whenever an accident occurs in a certain area, all enterprises in that area are forced to suspend operations for rectification. When one person is sick and everyone has to take medicine, it leads to an increase in overall societal costs, which in turn affects the normal functioning of the economy, creates new conflicts, and results in what is commonly referred to as a \"**failure**\" problem. Furthermore, this high-risk characteristic also hinders safe production efforts from attracting and retaining the best cadres and talents. To avoid imposing regulatory measures without considering the social costs, our country should establish a reasonable performance evaluation system for safety regulation, thereby motivating officials to take into account not only the losses caused by accidents but also the costs associated with accident prevention when carrying out safety oversight. At the same time, as the responsibilities of cadres increase, it is necessary to enhance the benefits associated with carrying out safe production work, so that everyone realizes that although this work involves heavy responsibilities and high risks, it also offers good prospects for development, thereby achieving a balance between costs and benefits for cadres in their work. In fact, China’s laws and regulations on work safety clearly stipulate that enterprises are the primary entities responsible for work safety. The safety supervision authorities merely carry out inspections and provide guidance to ensure that corporate production activities comply with laws and regulations; they do not bear direct responsibility for accidents that occur. However, under the high-pressure environment imposed by accountability measures, *** staff operate with extreme caution, feeling extremely uneasy; there is no objective standard for distinguishing between proper performance and negligence, leaving them at a loss as to what to do. Some officials can even only rely on having a bit of luck during their tenure, which makes it difficult to carry out accident prevention efforts proactively and effectively. Therefore, in the current challenging situation regarding work safety, in order to hold ** and safety supervisors accountable, it is necessary to establish a mechanism similar to the \"Harden Rule.\" When assessing regulatory responsibilities, the duties of ** and supervisors should be linked to the efforts they make in accident prevention, thereby promoting the management of risks related to work safety accidents and the development of rational prevention mechanisms. Therefore, law enforcement oversight and penalties for dereliction of duty and illegal acts should be strengthened prior to the occurrence of accidents, with an emphasis on shifting the focus of accountability forward. Overemphasizing accountability is not conducive to uncovering the truth behind an accident and thoroughly investigating its causes. The main task of an accident investigation is to determine the truth by collecting various types of evidence, analyzing the progression of the accident and the various complex factors related to it; it is important to avoid external interference and influence. In fact, factors such as the threat of litigation, conflicts of interest, and pressure from relevant parties and public opinion inevitably affect the assessment of the investigation process and its outcomes, giving accident investigations a more political and moral tone; this even poses a risk to objectivity and fairness. As a result, it is difficult to draw proper lessons from these incidents, and it is hard to use such findings as a basis for making institutional improvements. In recent years, in the early stages of investigating major accidents and disasters, people have sometimes struggled to determine whether such incidents are events or accidents, explosions or fires, and natural disasters or work-related safety accidents. Sometimes, the stakeholders involved in accident investigations may have significant disagreements regarding the determination of the cause of the accident. One of the main reasons for these phenomena is the division of departmental responsibilities and the assignment of accountability to the officials within that system. In 2013, a fire broke out at a poultry company in a region in the northeast, resulting in severe casualties and huge financial losses. Preliminary investigations suggest that the fire was caused by an explosion resulting from an ammonia leak. Based on this determination, the relevant authorities have ordered ammonia-related enterprises to strengthen safety management, and in some places, they have even forced such enterprises to suspend operations for rectification. As the investigation progressed, it was determined that the accident was caused by a fire resulting from an electrical fault, with ammonia not being the main culprit. However, for some time after that, certain areas continued to carry out follow-up inspections of safety measures in ammonia-related enterprises; the result of such attempts to fix things after the damage has occurred might be that both the sheep are lost and the pen remains unrepaired. Methodologically speaking, restoring the truth of the accident and identifying its specific causes are the top priorities in an accident investigation. Secondly, legal accountability is pursued against the actual illegal acts based on the conclusions of the accident investigation. Another purpose of accident investigation, which may be even more important, is to transform insights gained from the identified problems, especially those related to systemic flaws, into knowledge, and then turn that knowledge into established systems. This approach helps to fill gaps and improve the system, thereby creating a long-term solution. On November 20, 1996, a fire broke out at the **Kali Building, resulting in 41 deaths and 80 injuries. It is worth noting that **after the investigation into the \"11·20\" fire accident, no individual was held legally accountable; instead, the causes of the accident were attributed to a series of problems within the public safety systems and mechanisms in place at the time. It was concluded that the root cause of the accident lay in structural defects, which led to a range of institutional improvements in the field of public safety, including firefighting services. These efforts have significantly improved the ability to handle major and severe accidents. Another drawback of excessive accountability for accidents is that it can mislead the public and the media. Since production safety accidents are all characterized by inadequate assignment of responsibilities, they are classified as liability accidents; this may lead the public to believe that such accidents are related to poor performance and dereliction of duty on the part of officials. This tendency to think in such a way will continuously undermine the credibility of the grassroots level, and it may also lead the public to believe that preventing accidents and providing emergency rescue are responsibilities of that level alone, with little relevance to them. These two potential cognitive tendencies are highly detrimental to promoting safe production practices throughout society as well as to the development of public safety. IV. To remedy past mistakes, it is essential to rely on sound systems, apply measures with appropriate severity, and ensure clear rewards and punishments. The efforts in our country to implement safety production responsibilities and to hold those responsible for accidents strictly must be strengthened, not weakened. However, only by continuously reflecting on and evaluating the implementation of the system, identifying existing problems in a timely manner, and making ongoing improvements can the system be further refined, resulting in better implementation outcomes. Over the years, significant changes have taken place in our country’s economic and social conditions, and some of the provisions and requirements outlined in Order No. 493 are no longer applicable today. There is an urgent need to make certain modifications and innovations to the system for reporting, investigating, and handling work-related safety accidents. Against the backdrop of a severe current safety production situation and ongoing high risks of major accidents, improving the safety production responsibility system and strengthening accountability for accidents will inevitably become essential and irreplaceable measures in safety production efforts. One of the ambitious goals of our country’s economic and social development strategy is to build a moderately prosperous society in all respects by 2020, with simultaneous improvements in the situation regarding work safety. To achieve this goal, the implementation of safety production responsibility systems and strict accountability for accidents must be strengthened, not weakened. Our country is faced with a series of issues related to institutional and mechanism innovation, and institutionalizing the investigation and handling of accidents is an urgent priority. Practice in recent years has shown that strengthening accountability for work safety not only helps to uphold laws and discipline but also serves as a strong deterrent against those who are indifferent to work safety; it also inspires morale and enhances a sense of responsibility among officials and staff. A system or measure, while emphasizing normativity, must not overlook the guiding effect it has. Only by continuously reflecting on and evaluating the implementation of a system, identifying existing problems in a timely manner, and making ongoing improvements can the system be further refined, resulting in better implementation outcomes. At present, China’s work on work safety is faced with a series of issues related to institutional and systemic innovation, and the institutionalization of accident investigation and handling is an urgent priority. The pursuit of legal responsibility for production safety accidents should adhere to the principles of appropriate severity and strict punishment; strictness is needed to establish authority, while leniency must be based on trust. In fact, in our country, the pursuit of accountability and punishment for those directly responsible for production safety accidents suffers from both excessive severity and excessive leniency. As for the goals of reforming and innovating our country’s system for investigating and handling accidents, it is more about making such investigations and handling processes more scientific, effective, and credible, so as to enhance their role in serving as a warning, a source of motivation for improvement, and to further boost the credibility in dealing with major emergencies. There is an urgent need to revise and innovate the working system for accident reporting, investigation, and handling. To achieve this goal of reform and innovation, the author believes that the current priority is to amend the “Regulations on the Reporting, Investigation, and Handling of Work-Related Accidents”. Since its implementation in June 2007, Order No. 493 has played an important role in regulating the investigation and handling of work-related safety accidents in China, and has become one of the most fundamental systems for safety supervision. Over the years, significant changes have taken place in our country’s economic and social conditions, and some of the provisions and requirements outlined in Order No. 493 are no longer applicable today. There is an urgent need to make certain modifications and innovations to the system for reporting, investigating, and handling work-related safety accidents. The author analyzed the problems present in Order No. 493 and believes that the following amendments should be made: First, Order No. 493 fails to clearly distinguish among three distinct legal entities—the parties involved in work safety accidents, work safety managers, and supervisors responsible for enforcing work safety regulations. As a result, it becomes difficult to determine each party’s respective responsibilities during accident investigations and handling; all such parties are simply referred to as “persons responsible for the accident.” In fact, this is equivalent to mixing up athletes, referees, and coaches (managers). This phenomenon suggests that China’s safety supervision system should further clarify its role as the entity responsible for **administrative law enforcement, gradually reduce its managerial responsibilities, and promote the modernization of this supervision system. In recent years, we have often emphasized holding those responsible for negligence to account. The most crucial aspect of this policy is to clarify what constitutes \"negligence,\" and systems should be established so that departments responsible for safety oversight have clear guidelines to follow. In recent years, some comrades have wished to emphasize duty fulfillment and exemption from liability. In fact, liability for negligence already implies exemption from liability for due diligence, but if only emphasis is placed on exemption from liability for due diligence, it may give rise to some new problems. The author believes that it is important not to fail due to the system. Second, Article 3 of Chapter 1 of Order No. 493 stipulates that accidents are classified into four categories—extraordinarily serious, serious, relatively serious, and ordinary—based on the casualties or direct economic losses caused by the production safety accident. Such provisions sparked some debate during the drafting and discussion of Order 493. At that time, some experts believed that direct economic losses and casualties were not homogeneous events and could not be treated as one and the same. In the new Work Safety Law, China’s work safety policy has been revised to read: “Work safety efforts should be people-oriented, adhere to safe development, and follow the principle of giving top priority to safety, focusing on prevention, and adopting comprehensive management approaches.” If direct economic losses are also used as a criterion for classifying accidents, it could lead to the mistaken perception that property and lives are of equal importance, which may result in a negative impact on the system. Using direct economic losses as one of the criteria for classifying accidents reflects traditional thinking and is no longer appropriate in light of modern Chinese social values as well as current policies and regulations. Third, according to Article 29 of Order No. 493, the accident investigation team shall submit an accident investigation report within 60 days from the date of the accident ； Under special circumstances, with the approval of the people’s **responsible for investigating the accident, the deadline for submitting the accident investigation report may be appropriately extended, but the extension shall not exceed 60 days at most. Compared to international practices, this requirement is too strict. Setting strict time limits on the investigation process not only puts significant pressure on the accident investigation team but also inevitably affects the quality of the investigation, which may ultimately impact the scientific validity of the findings. Foreign investigations into major accidents generally do not have strict time limits; our country should establish more scientific time limits for such investigations. Fourth, in accordance with Order No. 493, accident investigation teams in our country are primarily composed of officials from relevant **functional departments and industry experts. There is a lack of a strict conflict-of-interest avoidance system in the accident investigation process; many of the departments involved in the investigation are likely to be the same departments that bear responsibility for the incident. This approach often makes it difficult to ensure the objectivity and neutrality of the investigation results, giving the impression that the referee and the athlete are one and the same. The investigation and handling of accidents must adhere to the principles of transparency, fairness, and broad credibility. If the investigation results are not accepted by society, it may not only lead to greater economic and social losses but also affect **reputation**. To ensure the fairness, transparency, and credibility of investigations into major accidents or incidents, foreign countries often establish specialized accident investigation and analysis agencies in accordance with the law; these agencies bring together relevant experts and representatives from various sectors to conduct independent investigations into such accidents. Fifth, it is recommended that some standardized and practically applicable procedures be added to the amendments of Order No. 493. These include the division of responsibilities among investigation teams, steps for on-site investigations, requirements for evidence collection, qualifications required for technical analysis units, methods for victims or their relatives to file claims, the initiation and communication regarding judicial intervention, third-party supervision and scrutiny during the accident investigation process, as well as provisions for public disclosure of the investigation findings and communication with the public. For example, regulations on how to manage media interviews and public opinion guidance can be added, with particular emphasis should be placed on rules regarding the Internet and new media. In the process of accident investigation, it is recommended to assign a dedicated person to monitor public opinion. To various societal concerns, we should provide answers based on facts and make responses grounded in scientific evidence. Sixthly, it is recommended to pay special attention to feeding back the lessons learned after an accident investigation and applying them in practice as soon as possible. We must be good at learning from accidents; the ultimate goal of accident investigations is to enable our society to learn from them and avoid repeating similar mistakes. Since taking independent responsibility for investigating transportation accidents in 1967, the U.S. **Traffic Safety Board has issued 12,000 safety recommendations to 2,200 organizations and individuals. Of these, 58% were directed at the U.S. transportation agencies, requiring them to explain the actions taken on each recommendation within 90 days; if certain recommendations were not adopted, they had to provide reasons for that. After drawing conclusions from the accident investigation report and putting forward specific recommendations, the relevant **supervising authorities or regulatory bodies should oversee the enterprises or organizations in those sectors to ensure that these recommendations are fully implemented. Only in this way can the painful lessons learned from various accidents be transformed into a driving force for societal progress. Good rain knows its season; it falls in spring. Our country is in an important period of economic and social development, and innovation in the systems and mechanisms for ensuring work safety is on the rise; therefore, we have reason to expect more advancements in the reform and innovation of the systems used to investigate work safety accidents.