Actuarial Education

100% success rate ! Perfect rewards for my effort & I am proud, Navneeth -in the first attempt itself passing out with flying colours in CT5 from IFoA, London Sep'16 diet! There will be more......

A very interesting diet of exams coming to an end, soon. Hit-or-miss , only time will tell. Till then party time is round the corner ! :)

All the best to all my students both current and former who are writing the London IFoA exams now! "Vijayee Bhavo"

CT5 classes in full swing in Hyderabad, CT4 starting on 7th July, 2016. Aspiring actuarial professional can reach out (call/WhatsApp) to +919392472521.

To all my students and aspiring actuarial candidates. Upcoming Asian Actuarial Conference in Gurgaon in November, 2016. If possible please make time to network and enhance knowledge. Make use of early bird discount still available.
Check out the link below:- http://www.actuariesindia.org/subMenu.aspx?id=373&val=20th_Asian_Actuarial_Conference

Big Data Analytics Summit of NASSCOM in Hyderabad - a great opportunity for students to meet industry head. Registrations open at http://www.nasscom.in/bigdata/

Asian Actuarial Conference - should be a great place to be seen !

A lovely session by the Managing Director of Acted which designs the courseware for the actuarial students was held in the recently concluded 18th GCA in Mumbai. The most important aspects that decide the passing probability of the students elaborated by the MD: He had conducted a survey amongst his all tutors and listed the some of the following :
1. Start preparation at least 5-6 months in advance
2. Practice writing and more writing.
3. Solving past question papers are imperative
4. Most Important:- READ the question asked in the examination.

So everyone buckle up and make full use of the MD's points.

Big Data and shipping industry: A challenge and an opportunity
1 INTRODUCTION
Shipping is one of the oldest industries in the world, and maritime trade goes back to the earliest recorded human history. Over history, it has been dominated and funded by large nation states, who used trade surpluses to build their empires, as well as merchant traders to build their enterprises. Merchant shipping, which has been, and remains the backbone of commercial trade has however, always been vulnerable to the vagaries of economic cycles, and it has been always been the endeavours of both shipping lines and merchant traders to anticipate the trends and maximise their benefits from them.
The current economic cycle also presents some difficult challenges. The secular downturn in commodity prices, a slowing economy in and the crisis in Europe has created headwinds that the rest of the world has not been able to absorb.
However, modern technology and advances in academic research in the past decade have started providing new insights and competitive advantages to industry participants, and newer ways to profit from its benefits.
Now we can and will address one specific trend in the industry today, and try to analyse its implications and opportunities.
2 CHALLENGES FACING THE INDUSTRY
A Nov 2014 study by McKinsey, the global consultants, surfaced challenges to the shipping industry which include
• Market saturation, and the resultant race for market share at all costs
• Pricing at, or near the marginal cost of services
• The industry is not seeing enough innovation in creating and pricing new service offerings
• Network designs are not keeping pace with changes in fleet structure
• Inherent conflicts between transporters and the owners of ships that they manage
These challenges are well known and well understood across the industry. However, opportunity lies in working around these challenges and if possible profiting from the new opportunities they create.
3 WHAT ARE THE EXISTING ASSETS THAT CAN BE LEVERAGED?
The shipping and transportation industry always had an acute understanding of its chief assets viz. the ships, vessels and the portside as well inland infrastructure to manage the transportation to and from the customers’ premises. However, modern research has shown that there are two other classes of assets of assets that have always been grossly under-utilized. They are both “intangible” as well as “non-monetized” but as experience across other industries have shown, they can nevertheless provide a decisive competitive advantage to any organization who can use them well. These are
1. Data- both internally generated data as well as externally “procured” data
2. People- who carry the tacit knowledge of trends, ground realities as well as the institutional memory that can serve the organization well, if combined with insights from data in a structured manner
Let’s analyze each one of these two assets in detail, for an organization needs both of them, woven together in a holistic strategy to get that competitive edge.
4 DATA AS AN ASSET IN THE SHIPPING INDUSTRY
Increasing automation and computerization is resulting in huge amounts of data being generated within any modern enterprise. Moving well beyond the traditional realms of financial management, companies are increasingly using data generated from every vehicle, aircraft, industrial systems and supply chain components to get new insights and gain a decisive competitive edge. Tesla Inc. is showing the way with connected cars, a concept of managing vehicles (and fleets) through data that is now increasingly disrupting the automobile market. Companies like Boeing are transforming into “aero-health” service providers that provide customer (for a fee) near real time, fault analysis, performance monitoring and customized alerts, allowing them to optimize their fleet operations in a significantly more agile way.
A modern shipping organization generates a large volume of data from multiple sources. Some of these sources include:
- Telemetry data from the vessels themselves. While the older vessels may not be fully equipped, many modern vessels come with a capability to capture extensive telemetry data from a range of on-board systems and instruments
- Operational and commercial data. These may come from onboard systems, as well as other related operational systems relating to energy consumption, port-side operations, cargo operations, manpower planning.
- Financial data. This is the traditional source of data for most analyses. However, modern technologies permit more extensive data capture as well as more insights.
The deluge of data, and the techniques required to analyze them, need a different approach, that of Big Data and Big Data Analytics.
5 WHAT IS BIG DATA?
Broadly, “Big Data” is an accumulation of data that is too large and complex to be handled by traditional database management tools.
The most common definition is the 3 Vs of the big data: Volume, Velocity and Variety – originally strung together by Dough Laney of Gartnet to which IBM introduced the fourth “V” viz. Veracity.
Volume: The sheer volume which often these days runs into petabytes poses technical challenge in terms of storing, sorting, processing as well as analyzing.
Velocity: In many cases in today’s world making the decision at the right time is of foremost importance; who wants to cross the road using data that is 2 minutes old? To stop credit card fraud, the decision has to be made before the transaction goes through. Tackling credit card fraud, the decision has to be made before the transaction is committed. Handling such problems in real-time adds further challenges to the system.
Variety – Data can be of any type- structured or unstructured, text, time series from sensors and log files from web sites or even position data from GPS loggers to list merely few. The challenge lies in combining the various data and to continuously being able to integrate new data as it becomes available.
Veracity highlights the ability to trust the data. Mere access to massive datasets does not guarantee right answer, having the wrong or biased dataset even if large will result in wrong answer. Trusting the quality and source of the dataset is thus an important aspect which becomes an increasing challenge with growing size and complexity. This highlights the importance of data governance.
In a nutshell we have a big data problem when the current (standard) toolset limits what we can do with the data and a new tool and/or algorithm is required to solve the problem.
These definitions are relative, and what is currently considered as big data will change with time and technology’s evolution. Nevertheless, these definitions are quite useful in terms of demystifying “big data” and making the technical aspects more concrete.
6 HOW CAN BIG DATA BE PRODUCTIVELY USED?
Storing, and “harvesting” big data is evolving into a mature discipline with a base of technological solutions that is both evolving, and maturing. Hadoop, and its various flavours have enabled organizations to store and get insights out of large volumes of data. Modern “cloud-based” implantations of Hadoop by large vendors like Amazon (“S3”), and Microsoft (“HDInsight”) have further reduced cost of implementation and lowered the barriers to adoption for these cutting edge tools. However, to businesses, the crux of the matter still lies not just in the technology barriers that need to be overcome, but in issues more fundamental. Organizations are still struggling to address how they would leverage the conclusions and insights from these Big Data analytics projects, and how to weave them into day-to-day decision making processed.
This paper will deal with some aspects of Big Data implementation and consumption within organizations.
6.1 “TOP-DOWN”- START WITH THE PROBLEM.
It is easy to be tempted by the host of technologies, solutions, and the benefits they promise. But it is always important to start at the beginning by “defining the problem statement”. This links all investments in Big Data analytics to tangible business outcomes.
Luckily, traditional six sigma provides an excellent (and well proven) framework for addressing these. Both the traditional DMAIC (“Define-measure-analyse-improve-control”) methodology as well as the DMADV (“Define-measure-analyse-design-verify”), also known as Design for Six Sigma (“DFSS”) provide an entry point for bringing statistical analytical tools at key stages of the project. While current implementations across the world depend on traditional techniques, this construct can be leveraged to both engage with, and leverage the output of Big Data analytics tools and techniques. This will not only re-envision lean six sigma for the modern enterprise, but it helps big data analytics becomes a tool to achieve larger business goals in supporting the measurement as well as analysis of alternative hypotheses and solutions.
6.2 “BOTTOMS-UP”- START WITH THE DATA, EXPLORE ACTIONABLE INSIGHTS
An alternative approach is to start with the data itself, while remaining agnostic to the insights (and surprises) that it may throw up. This involves a teamwork between data scientists/ specialists in big data analytics techniques and employees who specialize in their own domains. This team collaborates by contributing their respective areas of expertise. The data scientist creates the data models, “manages” the data flow, and refines both the model and the data sources. The business expert helps “sanity checks” in the way the data is used, in interpretation of the findings and most significantly in determining whether the finding is relevant to follow-on business decisions.
However, in real-life, a mix of both a top-down as well as a “bottoms-up” approach may be necessary to maximize the opportunities presented by this new technology.
6.3 TRAINING THE FRONTLINE STAFF
Just hiring data scientists and building models are necessary, but not sufficient preconditions for success. An equally important element is the training of frontline staff. Staff, and line managers need to be trained to “consume” the output of these models, have the necessary skills to interpret the results and finally, have the acumen and vocabulary to engage with the experts to provide feedback to continuously improve the effectiveness of the model.
A trained, “data-aware” staff, in collaboration with the analytics experts and data scientists, make the results of Big Data analytics come to life. They test the validity of business hypothesis, execute on strategies based on data models, provide feedback on success and failure and most importantly, form the most crucial link between Big Data analytics as an academic exercise and as an instrument of business decisioning.
6.4 ENGAGING THE ORGANISATION
Engaging the broader organization, and helping create an organizational culture that is data driven, and is willing to stake business strategies based on assumptions based on “mathematical” models. As with any new, transformational initiatives, profiting from Big Data is more likely to succeed (or fail) due to an organizational culture that embraces the opportunities that it represents, or resists from insecurities real and imagined, and cause it to fail.
Success will depend on leaders, and key managers to lead and role model the right behavior that managed by data, training for front-line staff to enable them to appreciate the potential, and incentives and metrics that can showcase progress and success (or failure) in an objective way.
7 IMPORTANCE OF ACTUARIAL SKILLS IN BIG DATA STRATEGIES
Actuaries bring to the table a thorough grounding of mathematical and statistical sciences, and a fair degree of business acumen. Their unique skills can have a special place in any organisation’s Big Data analytics strategy. These include
- “Bridge the gap” between problem definition and the analytical model
It is an easy trap to fall into- for the business teams and the analytics experts to work in their own silos. However, the best outcome is a result of finding a high-impact business problem, and then marrying the right data source to an appropriate mathematical/ statistical model that can be “trained” to address this problem. The actuary, as part of the project team, can help bridge this gap. Their mathematical training as well as an appreciation of applying models to real would business problems is a key competency that can go a long way in ensuring a successful outcome.
- Capacity building within the organization
Wide adoption is the key to success in any initiative. In Big Data analytics, this involves building some amount of organizational muscle in structured analysis, using statistical techniques. It is not designed to create an army of analytics SMEs and data scientists (though that would be welcome for sure) but to create a broad, collective, capacity to analyze data, given certain results and conclusions from models, and in some cases, to engage with the data scientists in the meaningful way to help build and improve models.
Actuaries are again uniquely placed to provide these training inputs and help being the broader organization to a common skill level that enables them to “consume and engage”.

8 CONCLUSION- STRATEGIES FOR LEVERAGING BIG DATA IN THE SHIPPING INDUSTRY

Step 1: Recognise data as an asset, inventory data sources, and repositories-
There are different frameworks that are readily available and can be used. E.g. Data Asset Framework from the University of Glasgow and multiple other frameworks from both academia and private advisory organisations. A review of this vast trove of data assets, often from sources that lay beyond the conventional domains (e.g. instrumentation telemetry) gives new insights on possibilities and stimulates new ideas.
Step 2- Link Big Data analytics programs to business priorities.
To be useful as a business tool, and not remain and exotic toy, this initiative has to earn its place by solving business challenges. This is where the application of business acumen and leadership buy-in is most useful. Will this program aim to improve top-line revenue, identify potential new revenue channels, reduce costs, improve productivity (e.g. energy costs, journey route plan) and utilization of constrained resources (e.g. ship-hours, crew-hours)? These need to be asked in clear, unambiguous terms.
Step 3: Build organisational muscle, work on building the right culture
A data-driven culture will need the support of the broader organization. Though it is tempting to start these initiatives as a “pilot” in a silo, the transition from a pilot to full-scale deployment is often the most fraught, especially in large organizations. In my opinion, it is always better to leverage existing programs and structures like quality, Six Sigma, continuous improvement programs, and co-opt them into using Big Data analytics as “just another tool”.
Training the front-line in using modern analytical tools and techniques, “democratizing” the availability of data and insights through self-serve tools like visualization tools (e.g. Power BI) etc. will convince them of their stake in the game. It will probably take more time, but will provide scale and broader acceptability with the front-line and make them a stakeholder in its success.
Step 4: Invest in the right technology
Last but not the least, it is important to invest in the right technology to enable the Big Data strategy and bring it to light. There are multiple solutions by different vendors, but a few key points in making the right decision are:
- The technology should support the type of data, as well the volume of data generated.
- Big data technology consulting skills are expensive. The solutions should be reasonably easy to support and train future staff on, once the implementation team has left.
- It should provide a mechanism to put the data in the hands of the staff who need it most. This may involve additional costs, but its critical if the fruits of Big Data analytics have to see wider adoption.

It was Saturday, 22nd August, amidst a very pleasant Mumbai weather, that saw students from all across the country, congregate in Andheri, a very popular suburb of western Mumbai – the occasion, 4th students’ connect organized by Institute of Actuaries of India (IAI). Students poured in from all the four corners of the country – Kolkata, Jalandhar, Chennai, Hyderabad even Jalgaon to ensure a full house!

The session formally kicked-off by IAI president Mr. Rajesh Dalmia, started with young graduates and undergraduates brimming with both curiosity and questions.

The students’ “connect” started with Mr. Sunil Sharma, who is the appointed actuary and CRO for OM Kotak Life Insurance. While giving a snapshot of the Indian life insurance industry he illustrated for the students trends like those of savings’ pattern, insurance pe*******on and the growth in the number of players in the industry. All of these being relevant factors for the industry’s growth and those of the upcoming opportunities for actuaries of the future. The students were also enlightened about the number of “fellows” in other actuarial societies like IFoA, London and SoA, USA besides in IAI – both IFoA and SoA having more than 30 times the number of "fellow actuaries" in Indian socity! Clearly an opportunity in India for the meritorious students to go for! Upcoming areas for employment of actuaries of future were also listed, education, investment banking, ERM and IT being some.

Similarly, students were treated to an overview of the other industries by the panel -general insurance (by Kirti Kothari, Reliance Genenral Insurane), Reinsurance (by Chetan Toshliwal Munich Re), Risk Management (by Lee Waddle, Aegon, Religare), Outsourcing/ Pensions (by Prabhakr Veer) and Life insurance (by Avdhesh Gupta), which was very insightful for the students.

Final icing on the cake for the students came in the form of the “Examinations- How to tackle them” by Mr. Subhendu Bal (Appointed actuary, SBI Life) – a bone tickling video from movie “KICK” left the students both “tickled and charged-up” to face “the tiger” in upcoming exam diets in October, 2015.

The details of the session can be relished at leisure by visiting IAI’s website at: http://www.actuariesindia.org/subMenu.aspx?id=334&val=4th_IAI_Connect

4th IAI connect for actuarial students, Mumbai