Japanizing Nigeria's airlines: Let us share the risk!

So Nigeria suffers yet again due to another air crash disaster. Anger, grief, questions, claims, finger-pointing and shock has engulfed the nation. Although the questionable history and physical condition of the plane at the time of its purchase actually borders on the criminal (in my opinion), I shall not dwell into that. Let us rather consider that a retired Nigeria airways pilot had predicted two weeks before the crash (Vanguard, 21 May, 2012) that such a disaster was in the offing on account of the 'cutting of corners' and sharp practices that characterised many airlines. Capt. Tito Omaghomi stated that current pilots are 'forced' to report and enter faults on scraps of paper instead of the statutory log books. Such pieces of paper could then easily be discarded (without trace) if the observed anomaly may require expensive maintenance procedures.

The governor of Akwa Ibom (Gov. Akpabio) also called the Managing Director of Dana Air not long (weeks, maybe days?) before the crash and warned him about the air worthiness of their planes (Vanguard, 3 June, 2012). This warning came after the same ill-fated Dana Air plane developed a fault en-route Lagos from Uyo. The plane was said to have hovered above Lagos skies for hours before it could finally land in Lagos airport. Furthermore, staff of Dana Air have told Channels TV that the plane was forced to fly that day despite it being classified as unfit by technicians. All these events prior to the disaster of 3 June 2012, are precursors of long-overdue radical changes to the Nigerian domestic airline industry.

Lets take a look at how Japan was able to cope with a similar disaster almost 30 years ago.

When in 1985 a Boeing 747 owned and operated by Japanese Airline (JAL) crashed killing about 500 people, the incident not only shocked the Japanese to the core, it jolted them to strive even harder for perfection. Their national pride was at stake, the reputation of JAL as an efficient and dependable airline not only suffered, Japanese citizens also refused to fly with JAL. So after three different boards resigned (not to mention a suicide or two), the management of JAL came up with a brilliant air safety plan.

The integrated maintenance of all JAL planes was cancelled. Every aeroplane was subsequently allocated a permanent maintenance crew who knew the history of their aeroplane and were compelled to fly on it for every trip it made. This is a practical example of risk management. To further assure the public, the names of all the members of maintenance crew for each plane were made public. No hiding place. Nigerians can not only learn from this, but we can take it to the next level. I therefore propose the following risk management measures.

As a preamble, only brand new planes should be used for domestic operations. All airlines that have planes more than 20 years old, should have their licenses suspended until they meet the new criteria. Secondly, each commercial airline operating within Nigeria should be compelled to have a permanent engineering maintenance crew for each individual aeroplane. This would ensure continuity (and improve the quality control) of maintenance issues. It will also create jobs, by the way. Thirdly, 10% of the seats of each commercial aeroplane (and not exceeding 10 people for planes with over 100 passengers) must be reserved for the permanent engineering maintenance crew who MUST be accompanied by at least two representatives from the management side of the airline company. In other words, a flight carrying 150 people should (as an example) have 7 maintenance crew and 3 administrative staff accompanying the 140 passengers.

The rationale for this approach is simple: If an aeroplane is well maintained, the engineers and representatives of management should have no qualms escorting the passengers to their destination. In other words, let us share the risk.  And we should share this risk not just with the rank and file (i.e. technical/engineering staff) like the Japanese do, but also with management staff of each Nigerian airline. Subsequently when the pilots complain of a fault or the engineers request for new parts, the management will take such complaints or requests more seriously. Those who think this is far-fetched or extreme should tell us if the lives of maintenance crew or management staff are more valuable than the lives of passengers. Besides, the Japanese Airline (JAL) has been doing this since the mid-1980s and neither the sky nor their profits have fallen as a result. If we are serious about progress and safety, lets adopt this modified version of their technique. 

May the soul of the departed find eternal peace...and may the guilty be punished at the end of the Dana Air crash investigations.

Z. Aliyu
zed@poetic.com

Jegamatics (Part II)

Previously in Jegamatics Part I, we looked at a simplified model for Smooth Elections in Nigeria using some road-side variables. Those who are sworn enemies of arithmetic, may want argue that you cannot compute an event/variable like Ballot box snatching (Bs). Sorry, but actually, you can do so - using probabilistic techniques, just as we can compute the probability of an event, say, dying from a car accident.

For example, to compute the probability of someone dying in a car accident in a given state of Nigeria, we need some data like (a) number of deaths caused by car accidents in the given state, e.g. 437 deaths per annum (b) population of the given state, e.g. 1,300,000. Therefore the probability of someone dying in a car accident is given as 237 divided by 1,300,00. Which gives 0.000336 or 0.033 percent. With the right data, we can compute the probability of many things; not just a ballot box being snatched, but even the probability of rainfall, winning the world cup or having a set of twins in a family. Anyway, once the probability of ballot snatching is found for a state or locality, it can then be inserted into our Equation 2.

Back to Jegamatics and planning for elections in Nigeria. Now, there are other politics-dependent factors that influence Smooth elections, (Se), which we did not consider in Jegamatics Part I. These include INEC's Fairness to Political Parties (Fp) and Overheating of the Polity (OHp). Let us also introduce the almighty 'Rigging' as a new variable. And because almost every party can be assumed to rig in one way or another, (or has the potential to rig) we would call it the Rigging Constant (Rc).

As before, we can see that Fairness to Political Parties (Fp) is a 'good variable', while Overheating of the Polity (OHp) and the Rigging Constant (Rc) are 'bad variables'. So if we now use only these three variables to determine Smooth elections (Se), we would end up with Equation 3 as shown below:

Equation 3 is saying that Smooth elections, 'Se' is directly proportional to INEC's fairness (Fp) and inversely proportional to an overheated polity (OHp). In other words, for Prof. Jega to maximise the chances of Smooth elections (Se) he should increase the fairness to all political parties (numerator) and strive to reduce the overheating of the polity (denominator). Although, the model is also telling Jega that even if he does these two things, rigging could still have an impact (since it is a constant). Now is the time for readers to buckle up because we are about to start Jegamatics 102.

Recall that in Jegamatics (Part I), we created a simple model (i.e. Equation 2) for Smooth elections, Se. That initial model was given as :

But we have just used Fp, OHp and Rc to derive a completely new formula for smooth elections (i.e. Equation 3). Is this a contradiction? No! Actually, since both models are saying the same thing in different ways; this means both Equation 2 and the new Equation 3 are equal to each other. Meaning that they can be combined, to give us a more robust model as shown by Equation 4:

This is the part where it gets really interesting because if Jega wanted to calculated the likelihood or impact of rigging in a particular state - which is ruled by a notorious Rigging Party of Nigeria (RPN) - he would make Rc the subject of formula as shown in Equation 5 below:

It is also possible that if Jega and his team are interested in the level of Security lapses in a given state;  then SL should be made subject of formula. Again, I must stress that these models/equations are highly ( I repeat: HIGHLY) simplified versions and actual models will be more complex. In real life, mathematical operators like square root and exponentials nay be applied to get an accurate model of this sort of problem. But I don't want to scare or confuse some readers with such things as exponentials because X plus Y minus Z is already hard enough for some people. Additionally, because a lot of quantitative and qualitative data needs to be collected, continuously for a model to be accurate; it goes without saying that a model for 2015 may not be accurate for 2019. Modelling is a cyclic process which should change as reality changes, as captured in the Figure below:

It is based on the figure above that a model has been defined by Harper and Lim (1986) as a tool which reflects the actual workings of the real world. However, to get the modelling right, Jega needs to have capable hands around him. He needs mathematicians and statisticians; certified project managers; operations researchers, planners, economists, risk managers and security experts. He can even poach courier agencies to get logistics personnel – and not just surround himself with fellow professors of political science.

Such a team will give him the input he needs about what could go wrong, where it could go wrong and how it could go wrong. If a batch of ballot papers don’t arrive from country X to Nigeria four days before elections, what are the legal, economic, security and political implications; and (critically) what are the contingency plans, especially if the affected states are volatile? Based on sentiments, irrational comments by political god-fathers or past precedents, security-related questions need to be asked beforehand such as: where and when can violence be expected, who will be the likely target of such violence, how long could the violence last, where could it originate from; and of course, how can it be averted?  

In a manner of speaking, INEC under Jega should have a smart person whose job is to sit down in front of a huge map of Nigeria, surrounded by telecommunications devices, throwing popcorn into his mouth and wondering: ‘what could possibly go wrong in this or that state?’ Indeed a lot went wrong. Registration of voters took much longer that INEC thought it would take; ballot papers were delayed, some INEC staff were late, others were overwhelmed by sheer number of voters and sadly youth corpers were butchered in states that are known for butchery. Nobody prepared for these and excuses have been flying around, while innocent youth corpers are lying six-feet underground. And all for what? Because they wanted to serve their fatherland and help INEC do its work? I was once a youth corper: I could have been a victim.

So, going forward, we can either clap for ourselves, and wait for the next 4 years when these same problems will blight our elections again or we can (a) amend the highly inadequate electoral act, (b) initiate strategic planning at INEC and (c) enhance capacity and expertise of INEC staff. People tend to say that Nigeria’s problem is not policy making but its implementation. This is not true. A lot of our policies are crude and it is just that many people don’t know otherwise. Only a crude election policy would not make security of election officials a topmost priority. 

Now is the time for INEC to hire some bright chaps, send them to study logistics, operations research and so on. The skills needed are of higher level and cannot be obtained from one Kpangolo State University whose course are neither accredited by NUC nor up-to-date by global standards. If INEC has to ask a specific university to establish specialist postgraduate diploma courses or masters degree for its staff, so be it.

There are also experienced Nigerians out there with expertise in operations research, mathematical modelling and risk management. That they have not been co-opted into the system is where our leadership has not done well. Many of such experienced people work in manufacturing and construction industries all over Nigeria and abroad. The rest are wallowing away in our Universities waiting for the next opportunity to declare strike action. So, Prof. Jega, (as a former head of ASUU) please find these people and start using them now, even if you have to retrain them. Start making pilot studies and test runs during local government elections. Standardise the processes and procedures of resource deployment, voter registration and casting of votes. Write a manual for best practices in conducting elections - down to the ward level. Instil a culture of continuous improvement and learning in all your staff and establish a permanent team of planners and strategists.

Those who are interested in going deeper in operations research can look into topics like non-linear programming, decision theory, games theory, chaos theory, simulation as well as a topic called 'transportation and assignment problem' - which should be of particular interest to people in INEC.

In my opinion, Jega achieved what he achieved through sheer hard work even though questions remain about how free and fair the elections were. Going forward, it is time for Jega to shift into the scientific gear. Show Nigerians that we can do better by 2015. There is no area of expertise which you desire today, that you cannot find qualified Nigerians to employ, and the rest of the world seems to know this secret. We are Nigerians and this means we are capable of anything and everything. This I believe, from the bottom of my heart.

The rest is up to Jega and INEC.   

Suggested readings:
(1) Harper, WM and Lim, HC (1986) Operational Research; 2nd Edition, Longman: London.
(2) Hillier, FS and Lieberman, GJ (2005) Introduction to operations research; 8th Edition, McGraw-Hill: New York.