I Think We Are at a Point, Due to the Nature of Some of These Mishaps, and Through Reflection

I think we are at a point, due to the nature of some of these mishaps, and through reflection, that we know we are able to improve on the way we work and take decisions to improve both operations and safety, by using newer technologies available to us now, that allow better and faster communication and decision making. Some of us have been continuously bringing these points up in conversation in different lights, drawing on our experience and diversity. I think what is important to know and do to make constructive progress in this direction, is to map human expertise from the various directions into such systems that allow the systems to recognize, identify, signal, and distribute the parameters that potentially lead into risk, so that they may be handled partially by the system and totally by the experts, so that any developing risk may be handled via process to re-direct the operations into a manageable risk scenario. Although, many good steps have been implemented in certain capacities, it is clear, that there is plenty left to do and much is left to be desired in current configurations of realtime and collaborative decision processes.

We were discussing an example on another thread, but what is important to keep in mind here, is that we can only automate to a certain level, and that humans need to be ultimately responsible for resolving any risk or danger that develops. Technology can really only help us to not overlook an event that is developing on several fronts, where it would normally take more than one person to notice, by bringing the information together from the different fronts, monitoring the combination of parameters, so that when a risk or danger develops, it is automatically recognized by the system, and then the system generates and distributes alerts and alarms to the "team" , whether they are found together or are distributed, to handle the situation. There are scenarios where such systems can be very useful in gaining time, where otherwise humans may still be squabbling between each other and explaining to each other, the system may deliver an undeniable flag to wave. There are cases where you have total system takeover and action, such as in a fire extinguishing system, where not only the danger is detected, but the action decision is also automated. At the well level or reservoir level, much more complicated scenarios may develop, and as much is due to subsurface conditions that we cannot normally see anyhow, for many safety processes, we are far more dependent on system detection. It should be said, that regulators are a different kettle of fish than operators. A single operator through its' own experience may develop a process and system that works very well, but is not in the business of sharing this development. Regulators while visiting such an operator may realize, oh, we should be doing this everywhere, and then they can recommend or develop standards. I find that these standards and regulating bodies are very instrumental in helping to assure levels of quality on an industry-wide basis, but they will always lag a bit behind the operators as proprietary developments are often used competitively.

One common theme that is touched on, after reading through this thread, is distracted mainly by the "How?" and these are minor in my opinion. Once it is clear to the critical mass of people exactly what the most critical missing piece to assessment and communication of risk is, then, and only then, providence will move to make this happen. Garry was discussing, with me and many others, back a year or more ago as most present were in agreement of the missing piece and now Garry is bringing this notion back to discussion and it still isn't something everyone really wants to "stomach" because it is difficult and difficult things get "skipped over" until all of the easy things are proven to still lack the proper solution. The missing piece, in these most complex and dangerous drilling operations are ubiquitous access to summations of risk from individual "risk silos". Peter, back many comments prior, mentioned Montara and if anyone wants to go back to the Montara summary report produced and presented by PTTEP, it is easily presented that PTTEP agrees, and made recommendations and presented the need for a similar functioning system. Some on this thread are confusing Process Safety with Ella's original premise being Drilling Process Safety. While the concept is basically the same the unknowns in drilling make things more interactively complex as the earth model is tightly coupled with the containment system. This doesn't make things easier. Let's not redo the discussions that have already been done on that front. They are well documented and there is much written, by myself, on the subject. There are companies that have individuals that are sold on the concept and yet there are many more that have not implemented any system that accomplishes the lacking that many more see clearly exists. It is good to see Garry writing about the need for exactly this here a year later and frankly the state of the industry in these regards is exactly as anticipated. Every easy idea and schema will be developed before the more difficult needed ones will be adopted.

Ella let me answer your question. There are many very good comments that describe the process of assessing and accounting for risk and the proper procedures and need for diligence in doing so yet it is how the individual risk assessments and the "a priori" and "dynamic" risks are neurally connected to a "master" risk assessment in "real time" that is lacking still. The main reason is it is "too difficult". Look, the most important element in any enterprise is diligence and patience. All of the latest rounds of regulations, actually, are attempts to regulate "diligence" by prescribing actions that ensure it, yet there will always be "loopholes" in those strivings. Over the last year, I've had the great opportunity to work with many people and companies on exactly this subject and I've seen that almost everyone agrees that this is the one most important thing to lend the situational awareness, spot anomalies to pre plan expectations and ability to respond safely to dynamic conditions we encounter regularly. We do not know all companies do privately yet publicly the efforts to this point lack ubiquitous, dynamic risk level assessments and broadcasting. Also, one comment on the commercial motive that tends to haste; it must be controlled. The same sentence in a book of ethics and morality says, "diligence leads to profit and haste leads to waste and ruin", and we can all see that play out in most accidents from stumbling to get to a toy by a child and cutting corners to get finished in industries. Point is regulators have known this for decades and through many rounds of disasters and try to prescribe formuli and red tape to ensure diligence and prevent haste and yet a better way is to mix in "goal orientation" and the goal should be to alert to people at all times the danger level because people tend to understand and take more caution during times when a "red flag" is raised and it prevents complacency likewise during times of a "green flag". Also gathering this information supplies formality to the type of communication and risk assessments needed on a constant dynamic basis in any enterprise. This simply remains lacking in ubiquitous forms. My suggestions are to go back and read Garry's 1st and 3rd comment. Garry repeated his 1st comment because everyone skipped over what he was saying and the same for his third comment. My first comment is the same as what Garry is saying. My suggestions are to focus on that, eliminating haste and ensuring diligence.

Prescribing how this is done, once this is accepted and focused on, will lead many to understand the need for situational awareness provided by summation of risk silos. With knowledge of this people tend to self regulate their haste and motivate themselves to diligence focused on the most important thing at every moment in time.

Wayne, my heart goes out to you. You mention, "commercial interests" and yet lets distinguish that not all commercial interests are necessarily lacking and yet only when they detract from diligence and lead to haste. Okay, so once we agree that the main focus should be on ensuring diligence and eliminating haste, we might also note that by definition diligence means a lack of haste or acting before needed precaution, etc. so let’s focus on mandating diligence. There are companies that are doing very well in this regard on their own using screening tools etc. to identify complex wells and using checklists on this to ensure that due diligence is performed. While there is still room for cutting corners by individual people this is less likely and performance controls on individual task completion may also be utilized. In terms of mandating this, as of yet, the new CFRs and the Final Rule of the BSEE gets mixed reviews and perhaps a failing grade on a few items yet a passing grade on the well containment screening tool, etc. For example look at a regulation that would have ensured diligence on the Macondo incident. We might have many choices of specific guidelines to require due diligence yet let’s pick one that is most specific and say that a more diligent design, monitoring and assessment of the negative pressure test would have ensured that the engineer that would have designed the test would also have had to monitor the test, assess the pressures and ultimately certify that the test passed or failed. Right? That is diligence in a nutshell. Yet four years after Macondo the following regulations relating to that specific action read: According to the Final Rule the 30 CFR 250.423c reads (c) You must perform a negative pressure test on all wells that use a subsea BOP stack or wells with mudline suspension systems. The BSEE District Manager may require you to perform additional negative pressure tests on other casing strings or liners (e.g., intermediate casing string or liner) or on wells with a surface BOP stack. And 30 CFR 250.423c1 reads: (1) You must perform a negative pressure test on your final casing string or liner. So the mandate from BSEE doesn’t ensure diligence and yet mandates that the test be performed and without ensuring diligence the test itself remains dangerous. Also, there is no mandate on how to test the shoe as an operator may chose to test the liner without having tested the shoe first or simply placed a bridge plug on top of the shoe without any due diligence into how far above the shoe a bridge plug might safe to drill out not knowing if pressure from the shoe had leaked, risen, and brought the gas pressure below the bridge plug. Of course oral mandates might prevent that and yet the most effective way to manage diligence in this case is to require the test itself to be designed, monitored, assessed and certified by one engineer, that has immediate access to every bit of information, preferably the operations engineer assigned to construct the well. Why the resistance? No resistance? Then why the oversight? No, oversight? Then explain why it isn’t important to regulate diligence in these straight forward and less prescriptive ways that really are goal oriented regulations that are simply stating a goal, safe testing, and ensuring this happens. Keep in mind that these aforementioned issues are in addition to the glaring lacking of ubiquitous "risk silo" summations. These same exact things are repeats from the last four years of discussions on these same topics. Not complaining yet let's keep it real.

Peter: your comment elaborates the daunting task of communication and competence assurance among teams that are "risk silos", if you will. Good thing to focus on as you obviously are well aware of. The most important thing to communicate is the key, in my opinion, and that is the risk level of each compartmental operation (risk silo i.e.. mud loggers, drilling crew, subsea engineering team, onshore design and operations teams, each piece of equipment or well bore barrier, etc.) to the maintenance of the barrier. Competence within each "risk silo" is assessed in the risk level. Communicating the risk level of each "silo" is achievable. Summing the contributions to the overall risk level that is communicated back to each "silo" is true communication and most helpful to situational awareness in which competency levels can then be tailored to match the current risk level, within each individual risk silo and triggering higher competence overall supervision and decision support levels as well. Otherwise, when the least competent individual is on the hitch during the highest risk you will see why his/her decision will be subject to higher probability of error. In a word, not tying competency levels to dynamic risk levels is "stupid". It should be done before the project begins so the vacation and well control school schedules of all can be synchronized with the pre planned estimate of the risk levels during the project. The Actual vs Pre Planned estimate make a good anomaly detection device as well.

Human Factors is classified by experts under the term Ergonomics, that is known by lay people as posture at their desk and how well we see our computer screens. Examples of ergonomics at work is using the term "Cockpit error" to replace "Pilot error", since a study of WWII pilots concluded that a small change in the position of an instrument in the cockpit of the air forces best pilots was causing them to crash. Similarly, the focus on giving individuals vital information in ubiquitous fashion, in our case, the current level of risk, is akin to having a well positioned gauge on an instrument panel that is essentially a "decision support panel", that all of us on the team are technically capable of understanding. Ella: I'm not a fan of a bridging document in the same breath as a ubiquitous all encompassing barrier risk level. A bridging document usually is only a statement of assignment of responsibilities and a risk level is an assessment of the current condition of the barrier. You and others have mentioned Montara root causes and I mentioned that the PTTEP final report had noted the need for a ubiquitous summation of risk silos in their report. Here is the link to the report: The line of sight tool is their version of the "master" risk assessment of the barrier I'm referring to so it cannot be said that envisioning the need for this is something beyond any one company that needs to be developed as a "cross-industry initiative". Let it be clear that in discussions independent of the Montara report and before any knowledge of their conclusions that the need for a ubiquitous summation of risk level of the barrier was noted as a primarily vital one. The PTTEP report simply became another confirmation of the same conclusion that almost every report hints at yet doesn't quite hit the nail on the head. Of course "In all labor there is profit, But mere talk leads only to want", so saying this is needed is different than actually doing anything. There are company(s) that have looked at the complexity of such a system and abandoned its development due to difficulty. That is human nature to abandon difficult tasks unless they are clearly recognized as being vitally necessary; a vision is created. Many times this isn't recognized until all the less vital, and difficult ideas are developed and found still lacking the affect of the more difficult and vital one. Clearly the "how?" is the needed process and is the difficulty and "why" is the needed outcome. The difference between diligence and haste is that diligence focuses on the process and the outcome, both, while haste focuses only on the outcome. That sole focus on outcome causes accidents. Performance control in commercial businesses management theory are usually only outcome based and in drilling this is the days versus depth curves. A "balanced" performance control in engineering management must include metrics for the process and the most vital of these is the health of the barriers since the process of drilling itself is simply the removal of natural barriers to flow of hydrocarbons and replacing them with man made conduits of flow control. So with that in mind the risk level before spud is essentially the risk of the natural barrier failing just under the rig, or practically zero barring a major earthquake and the earth opening up precisely there, yet once the well is spudded in deepwater drilling riderless, for instance, the risk level grows steadily, drops, increases, stays the same, etc. over the project timeline. This is a process performance metric and not an outcome performance metric. Just the term "process" is a step forward, however, since it signals that focus on "outcome" must be balanced with a focus on "process", yet "outcome" is equally important and in our case here the outcome we need must not be abandoned simply because the process is too difficult.