• Riskope International

    Risk and Crisis Management Decision Making Support Tolerability and acceptability definition Coaching and Skills development
  • Risk and Crisis Management Decision Making Support Tolerability and acceptability definition Coaching and Skills development
  • Archive

  • Categories

  • Meta

  • Flickr Photos

    Economic downturn crisis forecast November 2008

    Contact us to know details on economic downturn crisis forecast

    graphic results of economic downturn crisis forecast November 2008

    Economic Downturn Magnitude and Duration Quantitative Study by Riskope (http://www.riskope.com), November 2008

    More Photos

Japan tunnel disaster flashback

The very recent tragic accident in the Sasago Tunnel in Japan brings memories of our first mission in that country, in the aftermath of the Toyohama Tunnel accident in Furubira Town (Hokkaido, Japan) on February 10, 1996 (Toyohama Tunnel Rockfall).

When there is a probability of having people trapped underground, communication is the start and the end of a Properly done Risk Assessment

When there is a probability of having people trapped underground, like in any other accident, communication is the start and the end of a properly designed crisis plan.

The collapse of Toyohama tunnel crushed cars and a bus, killing 20 people in 1996. The causes of that failure were very different from the present case, but that’s not the point to be discussed here.
At that time, local authorities and emergency crews were heavily criticized by local population, and internationally, for the lack of “people-skills”, public relations, issue management and an apparent lack of empathy. Shortly after, we received an invitation and we were touring Japan lecturing on disaster communication, crisis management, with particular emphasis on the need to communicate with the families of potential victims and the media. That case study constituted the core of a course we gave then, within the frame of continuing education, at UBC, University of British Columbia.

Reportedly the Sasago tunnel is the worst such accident in Japan since 1996 with at least nine victims by the time we publish this post. We are going to follow very closely the development of the Sasago tunnel accident rescues hoping to learn that there are no more victims and the survivors’ families are  treated with compassion, empathy and their needs of information are addressed timely and in respectfully maner.

Financial Impacts and Risks Due to Sick-Leaves at a Swiss Luxury Watchmaker Factory.

The CFO of a client of ours, a Swiss Luxury Watches company, called the other day in a panic telling us: “I have heard from a guy who works in an international organization in Geneva which deals with Health around the World”, he said, “that this year is going to be the year of a severe flu pandemic. I need help in evaluating potential financial impacts due to sick-leaves”.

It was not to us to discuss whether the scenario of a major pandemic would be more likely this year than it was last year, or if next year will be higher or lower: we were asked to give some help in evaluating potential impacts of a scenario our client had defined.

The client had pretty good ‘historical’ data on duration of sick-leaves and number of simultaneous leaves during prior epidemics. None of these were, however, ‘complete’ statistics, so we had to widen the estimated ranges, develop a ‘comfortable’ set of probabilistic sick-leaves duration and simultaneous leaves estimates.

We explained to the CFO that data were clustered, showing indeed three ‘populations’ of sick-leaves: 1-6 days, 10-23 days, and finally more than 40 days…(including extended leave of absence or even death) which would of course have to be treated separately.

The CFO had already developed the ‘cost of consequence’, i.e the money-amount that a leave would cost to the company per day. Of course, the impacts varied as a function of the skills and the position of the worker. He had assumed the cost per day would be independent from the duration of the leave. Nevertheless it was a deterministic ‘one number’ evaluation, so we had to work with him to widen the ranges and develop a probabilistic per day financial impact resulting from sick-leaves.

Now it was time to evaluate risks for each ‘population’. See, neither the probability -p- nor the cost of consequence -C- are ‘one number’: both of them are affected by uncertainties, hence both of them are stochastic variables. Risk, expressed as a function of p*C, is therefore a stochastic variable as well.
You have certainly noted that we have made no assumption on the distribution of the variables. do you know why? Well, it’s simple! We do not need to assume any distribution because the calculations can be performed directly, with simple formulas! In this case no need to perform complicated analyses or to use Monte Carlo simulation: the results pop up from very simple straight forward formulas defining the average of the product knowing the average of the variables, then the standard deviation of the risk, knowing the variability of -p- and -C-.

At the end of the day the CFO had a clear understanding of the likelihood of sick-leave risks exceeding a value he had considered as the highest tolerable for the company.

He is going to develop, together with his Team, mitigative actions of various kinds.

The flu will not jeopardize his company, because he is now capable of “sizing” the problem!

Missing the “big picture consequences”

According to new data from the study “Sentieri” of the National Institute of Health, submitted to the Ministry of Health, in the city of Taranto, Italy, there was a 10% mortality increase in the period 2003-2008 with respect to the general Italian average rate. Taranto is heavily contaminated, in particular by a large foundry.

Classic forward looking” risk assessments missing the “big picture consequences”

Classic forward looking” risk assessments missing the “big picture consequences”

The trend reportedly confirms previous analyses, covering the period of 1995-2002, which studied the mortality profile of the resident populations in sites belonging to the national list of highly contaminated sites (about 60 in Italy), awaiting environmental remediation (the Italian equivalent of the US EPA Superfund list).

These sites also reportedly have a higher death rate, incidence of lung cancer in both men and women and higher than normal death rate for acute respiratory diseases (ARD).

From a toxicological point of view we are seeing and measuring here the adverse effects of long-term contamination on public health. As we learn immensely from failures, it would be useful to our Human society to see an unbiased toxicological risk assessments of Taranto developed a posteriori. As toxicological risk assessments are specialists approaches geared toward ascertaining adverse health effects from exposure to agents (in this case man-made contamination of the air), such a “back-analysis” would offer a platform to recalibrate how this type of assessment is performed, so that large projects (power plants, chemical plants, new chemicals to be released in the environment, etc.) will receive in the future better evaluation. At least, if we do so, all the pain and the suffering endured by the exposed populations will at least be used for a positive purpose.

If a toxicological risk assessment is performed on a generic contaminated site it typically includes an estimate of the probability of harm, such as the probability of liver toxicity or the effect on wildlife, human health and a clear description of the various assumptions and many uncertainties that go into the risk assessment. Like any other human endeavour, toxicological risk assessment may be biased, look only at “rosy scenarios”, or “average scenarios”, etc. The proposed Taranto back-analysis will help “clean-up” optimistic biases.

Let’s now talk about “classic forward looking” risk assessments.
Their goal is to provide decision makers, for example, government regulatory officials, industry health and safety directors, public health officials, with a basis for decisions making related to the use or the production of physical agents/contamination in order to protect people’s health and the environment. The decision-making process often involves factors in addition to the risk assessment results, such as social values, technical feasibility and economic factors.
As described by the National Research Council of the National Academy of Sciences, toxicological risk assessments have four components:
1) Hazard identification (what the agent/contamination “can do” to the exposed population, i.e, for example, the capacity to cause liver or nervous system damage, cancer.
2) Dose response assessment (how much of an agent/contamination is necessary to cause a toxic effect, and a prediction of exposure levels with negligible effects).
3) Exposure assessment (how much of an agent/contaminant people might be exposed to under various conditions).
4) Risk characterization (i.e., what is the likelihood that there will be an increase in cancer in a population exposed to the agent/contaminant).
A toxicological risk assessment will also discuss, as mentioned above, in detailed manner assumptions and uncertainties, in order to enable decision makers to understand how “safe” the conclusions are.

Can you see the piece missing in here? It is so big, right in front of us….
We are missing the “big picture consequences”…
For one person that might get sick, we are talking about an entire family suffering, psychological pressure, depression, psychosomatic ailments, etc.
It is too simplistic to characterize the risk by the increase of cancer rates: what about the lost agricultural production from contaminated farmland, lost tourism revenue, what about swimming restrictions in the lake or sea water next to home, and not being able to eat the fish, drink the milk, etc.
The common reply to this statement is: “well, it’s too difficult to evaluate”.
A very good excuse, isn’t it? But remain only an excuse.
Methods to get that difficult job done do exist and we use them, for example in the formulation of the consequences of Cyber-attacks, large scale environmental damages. They are certainly not deterministic, and they are certainly not 100% accurate, but rather than wallowing in self-inflicted blindness, we much rather prefer to be generally right than precisely wrong (or ignorant).

Risk Managers are retiring. Corporations have to fight the brain drain or be faced with over-exposures to risks.

The retirement of a corporate risk manager might be the source of critical exposures. It is actually a significant risk per se: one with a high likelihood (everyone ends-up retiring), and potentially large consequences, which can only be mitigated with proactive actions, to be taken well in advance.

As a matter of fact, if nothing is done to prevent this drain, risk managers will often take knowledge and on-the-job skill-set home with them, including precious historical knowledge of the job. That loss will be compounded by the fact that “replacements” will not be trusted, be considered as “unknown new-comers”. Furthermore, the overall corporate “experience level” will go down, as the US Bureau of Labor Statistics (BLS) predicts that between 2010 and 2020 there will be a total of 54.8 million job openings, over 60% of which due to retirements or people permanently leaving the workforce.

Smart companies start planning well in advance of any key person’s departure, but one is left to wonder if the so called “great crew change” should not be taken as the opportunity to enhance corporate risk management systems.

In nowadays competitive corporate environment companies want someone to walk in the door and jump right in with minimal guidance and training. To do that, documentation, knowledge and the basis for prior decision making should be clear and transparent. Not a single ERM Risk Record we have seen to date has those characteristics, not even the spiffy ones based on smart looking ERM/RA softwares.

The very nature of indexed Probabilities Impact Graphs (PIGs), semi-quantitative evaluation, poor understanding of consequences will make the task of the coming generation even harder than it should be. Their chances of failure are very high, and they are not to be blamed.

Riskope’s ORE (Optimum Risk Estimates) has the power to make life a lot easier to the coming generation, and should be deployed immediately to ensure that “stories from the trenches” do not fall in oblivion.

Intolerable risks per divisional element

One of many possible Riskope’s ORE Risk Cockpit graphs allowing a complete understanding of the Corporate Risk Environment and Landscape, a very valuable addition to classic ERM.

ORE, once deployed, offers a clear vision of how the business works and the resulting risk landscape, and facilitates open and transparent discussions on the evolution of both the business and its risks. ORE allows to capture the knowledge of key people about to retire by preventive and gradual “brain dumping” and in the meantime becomes a knowledge-sharing system which clarifies how decisions were/are/will be made, something that rarely exists in the industries world-wide.

In a world dominated by qualitative approximations, behavioral biases and related “intuitive feelings”, rather than rational uncertainty considerations, ORE will allow businesses around the world to prosper with the new generation of Risk Managers.

The King is dead. Long live the King!

The devil is in the details.

We all love rounding numbers, i.e. “erasing the decimals” by pushing the value up or down to the nearest, or more pleasing, integer.
When we let our “esthetics sense”, or our will to influence our audience, dominate, we will merrily jump way more than the decimals and come out with “media friendly rounded” values such as “one million people came to the streets”.

Often the nicest numbers will also be the ones the farthest from reality.
Many risk assessments, especially if they are used to obtain approval of a project, or an alternative selection, will use the esthetic art of biasing reality with artificial, misrepresenting rounding.
How is it done?
Well, these are examples of great classic approaches:
1) they start by saying that because no one can know what will happen in the future (who knows, a meteorite could fall on our head…), they will limit their analysis to credible (they do not define credible, of course), average consequence (they do not define average, of course). By doing that, they avoid bringing into the analyses all sorts of embarrassing high likelihood/low-consequence and low likelihood/catastrophic-consequence scenarios (Fukushima anyone? Or BP).
2) The second step is to divide reality in discrete classes, let’s say five “steps” of increasing likelihood and consequence, with carefully selected arbitrary limits of the classes. I oftentimes see risk assessments where reality has been consistently biased to fit the purposes of misrepresentation.
3) The third step consists in making everyone to believe that for each (risk) scenario it is perfectly OK to forget any kind of variability and “stick” it in a cell based on “rounded” value of likelihood and consequence. At this point many decide to go even further away from reality by using indexes (like 1-5, talk about rounding!) instead than “real” numbers.
4) Finally, another arbitrary set of assumptions leads to draw “stepped” limits splitting the cells in “rounded” categories driving attention level and mitigation decisions.
I will let our readers decide how far from reality such an assessment can be and how wrong the resulting decision making could be, especially if the project/alternative is not a usual one.

Close Calls and Human Biases

Near misses receive different interpretations in different countries, cultural backgrounds. In some cases they are considered glorified achievements of intuitive semi-gods, in others just plain indicators of near failure.

When near misses become repetitive, it can be assumed they are the result of a systemic flaw, which will only require a small “twist of fate” to turn into an accident, possibly a disaster. In an industrial operation we know of, people became accustomed to a loud bang coming from a particular device until one day the loud bang became a monster explosion that levelled the whole operation. It was only after the catastrophe that those same people connected the dots and understood that that bang was actually not benign at all, indeed the symptom of a systemic flaw.

The world would be a better place if we stopped glorifying those very few cases where, against all odds, near misses turn out to be prestigious successes, while forgetting about the scores of failures due to “selective bias” of information hinting to possible flaws. We Humans are indeed apparently wired to:

  • See patterns that do not necessarily exist.
  • Poorly conceptualize long arcs of time.
  • Perceive what agrees with our pre-existing expectations, and ignore things that disagree with our beliefs.
  • Forget our losers and over-emphasize our winners.
  • Be better at processing good news than bad ones.
  • Be vulnerable to anecdotes that mislead or present false conclusions unsupported by data.

At Riskope we believe that our Mission is to support any Human endeavour with rational methodologies that allow bypassing these biases.

Hazards, Safety and Security Management of Mining (and other Natural Resources) Access Roads

Construction equipment on flatbed Hazards, Safety and Security Management of Mining (and other
Natural Resources) Access Roads

Off-site road transportation (access roads) to remote mining (and other natural resources) operations (public, private or semi-private) poses a series of unique challenges to people in charge of their Safety, Security and Hazard Management.

This course draws information from a number of real-life studies performed on this type of road by the authors (Riskope) and its Road Safety Division (Eurosain), and presents them following the logic of a Safety, Security and Hazard Management Report that could be prepared for a road that crosses a variety of terrains and is subjected to a wide array of hazards, considering, of course, a variety of climatic zones. The course is divided into steps, which correspond to those necessary to complete a Safety, Security and Hazard Management Report.

At the end of the course, the course proposes a catalogue of mitigations that can be integrated in order to enhance Safety, Security and Hazard Management along any (mining) road. These include both physical mitigations on the road/infrastructure, and personnel hazards mitigations, including a review of driver’s fatigue, high altitude effects, etc.
Technical Appendices deliver specific information that will help Road Managers to design specific signals and signage.
This course (Live webcast on 10, 11 October 2012) makes references to prior Edumine courses by F. Oboni and C. Oboni whenever necessary, but doesn’t require their knowledge and does not repeat any information on risk assessment, risk, or management.

Mitigation erosion control

Two blasts have rocked saw mills in B.C. in recent times. One in January leveled the Babine Mill (Burns Lake), the other in April destroyed Lakeland Mills (Prince George).

In both cases, unfortunately, there were casualties.

In both cases, the blasts were apparently due to sawdust.

In both cases official inquiries are still ongoing, but the media have been filled with hypotheses and discussions. Reportedly (Globe and Mail, April 27th) “the B.C. Government waited until the second catastrophe this week to issue province-wide guidelines, inspection regimes, deadlines and, possibly, new regulations”. Reportedly (same source) “when an explosion and fire tore through the Burns Lake mill, owners of other mills across the B.C. Interior -including Lakeland- looked at housekeeping and upgrades to address the potential risk (NB: improper technical language: should read potential hazard, the difference is important, see below!) of sawdust explosions at mills working with dry wood killed by pine beetles”; “safety experts noted the potential risk (NB: improper technical language: should read potential hazard, the difference is important, see below!) of sawdust after the January explosion at the Burns Lake sawmill”; “the industry has already been struggling with a massive shift since the mountain pine beetle began the widespread destruction of B.C. Forests”.

Also, the Globe and Mail (April 28th ) quotes Margaret MacDiarmid, the province Labour Minister, saying that “Burns Lake fire had appeared to be unique, pointing to the cold snap in January that had forced the mill to close its windows, increasing the hazard of a dust explosion”.

Out of respect for all parties involved, scientific rigour, we will not discuss the general foreseeability of such an event (Globe and Mail, April 26th reports for example: “in 2009 an inspection report found that Lakeland had not been monitoring worker exposure to wood dust…”), possible explosion triggers, possible preventative actions, post-catastrophe measures, possible responsibilities.

Instead we will focus our attention on some Risk Management points.

  1. Many industries, regulatory bodies, organizations keep confusing “risks” and “hazards”. As far as we know, no risks were formally evaluated for the mills, but hazard inspections, at best, were performed. To use a very simple language, “hazard is something that has potential to go wrong”, whereas “risk is the combination of a hazard with its potential consequences”. The confusion can lead to opposite “equally inappropriate” results:

    a) If the environment is overly optimistic looking at hazards is conducive to reject any preventative action on the basis of “it’s just a hazard like another…we live with them every day”…things will remain as they are until an accident happens.
    b) If the environment is overly pessimistic then excessive preventative actions will be taken, with potential loss of competitiveness.

  2. If risks had been evaluated (properly) instead of just looking at hazards, it would have resulted that a sawdust triggered explosion (due to any reason), and resulting fire, had potential to generate casualties, destroy the plant. Most likely anyone would have considered that event intolerable, and things would have been corrected long ago.

    a) Well balanced regulations are risk based rather than hazard based.
    b) Risk Based Decision Making (RBDM) is a discipline that warrants proper scientific approaches. It is not “improvised”, it requires skills. Methods should be “enforced” rather than resolve to knee-jerk reactions.

  3. When performing Risk Analyses for industries around the world Riskope is often confronted with “long chain” domino effects. From what we have read to date, the explosions’ root causes could both be “climate change” based:

    -pine beetle size and severity of the outbreak and
    -the cold snap

    can indeed be considered somehow linked to the global climate change.

    We do not believe any serious professional would ever be claiming that such a scenario (starting with climate change….down to saw dust explosions) could have been foreseen. However, if during a Risk Assessment site-visit significant volumes of sawdust would have been seen, a fire or explosion scenario would have been generated (not important to define the cause of the dust), potential consequences evaluated, etc. (see point 1,2 above).

    It is time industries start looking at their risks in proper rational ways. Risk Assessment techniques will pave the way to safety, security, competitiveness and long term sustainability.

    A sustainable industry is an industry that can keep working without a break. Avoiding catastrophic accidents is the first step toward sustainability.

Protecting Critical Infrastructure Against Cyber-Attacks

Riskope was invited to give a talk in Switzerland at the Geneva Dialogue 16th-17th of April 2012. That conference was geared toward Critical Infrastructure Protection, Resilience Enhancement, Strategies for the Future with participants from Canada, China, various European countries, Mexico and the USA.

Riskope presented “Risk Management: steering your organization through difficult times” (presentation available for free to registered blog users on demand) where we showed how standard heat maps, probability impact graphs and other indexed approaches (PIGs) fail to fulfil the needs of governments and corporations willing to properly evaluate and keep up to date holistic (entreprise-wide) risk assessments (ERM), optimize their risk exposures to gain critical competitive edge, and steer away from downturns or crisis.

A case study from real life was presented where it was shown how predictive knowledge about depth and duration of the economic recession  originally published by Riskope in the fall 2008 can be integrated into an ERM approach. In the same approach various cyber risks can be integrated as well, allowing a process of rational mitigative decision making to take place.

PIGs, as stated earlier in this blog and in other publications simply do not allow the depth, transparency, rationality and repeatability offered by Riskope’s Optimum Risk Estimates (ORE) approach.

Risk perception, corporate prestige, psychological factors, alibi and denial.

Roughly ten years ago we were invited by a very famous and prestigious European Railways company (NB: the company name is covered by confidentiality and we have also slightly altered the story to further protect the identities of the implied parties) to give a seminar to top management.

European Railways companies are “state owned” enterprises, thus their Boards generally have CxO who either are administration veterans or politicians. Men of great experience, but traditionally poorly inclined to follow the rapid changes imposed by our fast evolving world.

The subject of the seminar was, of course, Risk Management. We were surprised that our client would be ready to part from the traditional all-hazard management, the obsolete approach inherited from the “royal rail-roads” era.

Railway, road, communication, and water supply disrupted by a land slide (natural hazard)

Railway, road, communication, and water supply disrupted by a land slide (natural hazard)

However, we learned that the “old boys” of that particular administration had either retired or were promoted to higher functions, and the new CEO, was very eager to promote “new ways”, including Risk Management, in the company. We were very honoured to have him sit in the seminar.

What Happened?

We started sharing our ideas about our numerous experiences with US and Canadian Rail-roads Risk Management, Wharves and ship loaders risk assessments, large-scale logistic risks, and country-wide approaches. Our intention was to then expand on other horizons, ending our prologue to the seminar with a panorama of strategic and tactical/operational risks before entering into rational prioritization and Risk Based Decision Making .

The CEO, however, quickly erupted with a statement that sounded roughly like: “well, I appreciate that you come from the technical side (a disdain grin appeared on his face..) and like to talk about these “technical/natural risks” (disdain again…), but we are here to hear about business risks, strategic risks, high-level understanding (happy grin, now…)…”. Believe me, if he did not use the “Black-Swan”  stereotype, it was just because no one had invented it yet! Without losing emphasis he ended-up saying that at corporate level they would like to use modern approaches (he even mentioned an off-the shelf methodology based on cards…). They were certainly not interested in knowing about “technical details” such as probabilities and consequence functions, tolerability, etc.

Our reaction was simple.

We said: “Sir, as far as we understand your company’s business is running trains; freight and passenger trains.

We understand our mission is to support your business’ objectives by bringing in good risk management and decision-making tools.

Only a rational and unbiased approach will tell which ones are the “major risks” your company is facing. Intuition is not enough, and actually, it proves generally wrong.

Risks are defined by their probabilities of occurrence and their consequences, and should be compared to your company’s tolerability for rational prioritization, and that’s way too difficult for intuition.

Without trying to expect the conclusion of the study you need to perform a ranking of your risks, we can say that the “high-level risks” you mention depend on political decisions (shall we call them political hazards?). Political decisions are slow (some might even never happen) by definition, especially when they would hit a state owned corporation. Instead, natural hazards are blind, they do not need to be re-elected, and climate is evolving faster than we might think. That’s why we believe it is not rational to exclude natural phenomena from the analyses”.

After that exchange, the CEO left, and we finished the seminar.

What Happened Next?

We knew we would never work again for that company/board/CEO.

We assumed they went shopping for a deck of cards (hopefully they were not Taro cards…) and started discussing high-level strategic risks.

Not more than six months later a landslide (a natural culprit of the “technical kind”) took away one of the major double track lines in the country, and the company was ashamed by the media as they had to shuttle people on buses around the natural phenomenon. It took roughly another year for a similar phenomenon to occur elsewhere on the network, triggered, this time, by “anomalous rainfalls”. Remember, Black Swans were not invented yet!

No need to say, no one ever called us back, but, a few weeks ago, we learned by a RfP from that country that:

a) the Ministry of Transportation had “taken away” from the rail road company their Risk Management function (a loss of freedom, like being put under monitoring by a higher authority).

b) The Ministry of Transportation had put together a Risk Management approach that goes down to the single switch, single signal, single rockfall detail at country-wide scale (so much for looking into strategic, high level risks only…NOT!). Pretty much what we had done for many other clients, but was thrown away by the CEO.

c) The Ministry of Transportation had developed a pilot study and is now in RfP phase for a custom tailored computer application at country scale.

Conclusions

So, you might ask, what are the conclusions of this story? Well, here they are, under the form of simple questions based on the verb “to lose”:

  1. Why lose ten years of precious time?

  2. Why go deep into crises, ending up losing your freedom?

  3. Why lose loads of money?

  4. Why lose your sight, blinded by irrational approaches?

When you see it written like this it sounds so simple, so easy, yet, when it comes to actually doing things, few companies, few boards of directors, few individuals actually give themselves the means to embrace rational and scientific approaches to Risk and Crisis Management using clearly defined tolerability criteria, the only way to be a winner rather than a loser…We will not discuss banks and governments, but you can see from the media, the story is perfectly mirrored in those fields as well.

Evolution

Because all of this sounds more like the result of cognitive biases, and other psychological effects  impacting the board of directors, seen as deeply interacting constellation of individuals, we have decided to team up with a specialist of systemic behavioural facilitation and we will soon offer in-house or public integrated sessions.

During these sessions Risk and Crisis Management, Risk Based Decision Making will be discussed while fostering deep understanding of the group biases and behaviour when the group is involved into decision making processes and uncertainties. The goal?

Your success, your survival, your modernization, your openness, and at the end, of course “your happiness and well-being”.

Contact Riskope to discuss the details, book a session at your HQ, or organize a public engagement. We are active world-wide and have deep understanding of cross-cultural environments.

Follow

Get every new post delivered to your Inbox.

Join 1,487 other followers

%d bloggers like this: