What Business Students Need to Know About Information Systems

C O N T E N T S:

KEY TOPICS

Originally prepared in response to draft accreditation guidelines prepared by AACSB International, the article includes a compilation of the concepts that the authors believe to be the core information systems knowledge that all business school students should be familiar with.(More…)

Average starting salary for undergraduate Management Information Systems graduates…

Predictive analytics is not enough

April 26, 2015

Decision intelligenceFinanceIncludes Interactive Decision ModelIntelligence Augmentation (IA)Knowledge Management

The idea of predictive analytics can seem like magic: how, really, can a computer predict the future? Yet we’ve seen a lot of success based on this advanced technology in recent years, from Netflix to Amazon, Google, and more. These companies mine a massive amount of data every day for patterns, and it drives massive revenues.

However, for a widespread class of situations, predictive analytics alone aren’t enough. Consider the decision model below, which I introduced in my last post. The blue graphs on the right-hand side are based on predictive analytics, but they are only building blocks in the full model.  They are not enough on their own.

(Note: some browsers, such as Firefox, will display this less interactively than others. As you move levers, you’ll only see a response after you let the lever go)

Only by assembling predictive analytics into a DI model does this tech realize its full potential.

CLICK TO TWEET

As you can see, the blue graphs are predictions about how one thing causes another. But they’re only useful when we combine them with levers, and then link them together with outcomes.  The prediction is only part of the puzzle.

To understand this, let’s start with what’s going on when computers predict the future.   At first blush, the idea that a computer can do this can sound fishy. Yet computers are indeed good at finding patterns, especially if there’s a massive amount of data, and if a pattern holds into the future.  Under these conditions, then, in a sense, this pattern recognition can be thought of as “future prediction”. This is the assumption underlying production predictive analytics systems today, whether they’re used for determining a pattern of credit card usage that indicates fraud, the pattern of Amazon browsing and buying behavior that predicts whether you’re going to buy a particular new book, or the pattern of search terms you type into google that predict your likelihood to click on an ad and buy a product.

The problem with these kinds of predictions in isolation occurs when the situation changes, and the old patterns don’t hold. Notoriously difficult, for example, is stock market prediction in a fickle world with macro effects that aren’t captured in the data.  Nassim Taleb understands this situation best: as a derivatives trader and financial executive in the years leading up to the 2008 financial crisis, he was a front-row witness to the limitations of these approaches, and built a successful investment strategy based on the recognition that the most important events in markets are those for which patterns in historical data cease to hold.

Black Swan theory, as formulated by Taleb, captures the idea that a turkey, shortly before Thanksgiving, is exposed to a considerable amount of data indicating that his world is safe, and that he’ll be well-fed and housed.  But the turkey lacks a knowledge of the surrounding system.  This creates problems.

Before Thanksgiving, the turkey is lacking a systems model.

CLICK TO TWEET

The problem: because he only relies on historical data, absent a cause-and-effect structure of the world at large, the turkey is in for quite a shock at the end of the month.

The benefit of a decision model: context that transcends turkey thinking.

CLICK TO TWEET

Another limitation of predictive analytics in isolation: they make a simplifying assumption about the decisions surrounding them.   Use cases in which predictive analytics alone are successful are also those in which it is feasible to avoid complex modeling of the impact of different decisions based on the analytic result, or where the decision is so simple, that such a model isn’t required.  These are the situations where there is no Thanksgiving (are there Turkeys in Australia, or just black swans?) and historical data is good enough at predicting the future.  The “black swan” mistake to assume this is true more often than it really is.

Take fraud, for example.  If we detect a pattern of usage that might indicate criminal activity—whether using a credit card, telephone system, or on the internet—then what’s next? Fraud experts tell me that, in general, there are many choices, such as to hand off to a fraud team for further investigation, to automatically freeze the card, and more. Each of these choices has different costs and potential benefits. Yet many systems today we ignore these distinctions, apply a one-size-fits-all approach, and so our ability to detect fraud is falling behind.

We must be as sophisticated in our decision models as we are with our predictions.

CLICK TO TWEET

Customer experience management (CEM) is another example. Many companies use “churn analytics” or “net promoter score” analytics to predict the likelihood that a customer will leave for a competitor. Yet what then? I’m sure that a crisp $1000 bill to each potential churner would reduce their likelihood to leave, but of course that’s not cost-effective.

Cost and benefit modeling is essential to predictive analytics, yet is often overlooked.

CLICK TO TWEET

So this is an exciting time.  We’ve become awesome at constructing the machine learning and statistical building blocks of a decision model.  These are like great chocolate, and now it’s time to bake the cake: we must integrate  predictive analytics within a framework of complex systems modeling to build decision intelligence support systems to obtain the full benefit of this groundbreaking technology.

Let me help you do that! 

It is always a challenge for companies to keep their operational costs to a minimum and avoid the frustration that arise if they transfer any additional costs to their clients during peak seasons.Is your  company adequately equipped to navigate the challenges of  maintaining the balance between not losing face with its customers and at the same time not inflicting heavy costs unto itself? 

This article by Dr.Lorien Pratt, chief scientist at #Quantellia and valued partner to #InformeddecisionsAB provides a proof of concept demo to model a Decision making scenario that US Cable operators should adopt if they want to “Stay ahead of the energy cost curve” during peak seasons. Our take at Informed Decisions AB is that if this scenario worked for US Cable operators, it should work for you too.

August 20, 2015

Cable Operators (MSOs)EnvironmentModel walkthroughSimulationTelecommunications

Many companies are facing the prospect of steep increases in the cost of energy in the coming years.  In response, many are looking at alternative energy sources. However, navigating the transition to this new world contains hidden dangers, so an evidence-based modeling approach can make a big difference.  This article looks at this decision-making process through the lens of US cable operators, to understand the specific decisions they face.

Worldwide, the cost of on-grid, coal- and petroleum-generated energy is expected to rise rapidly, while the availability of renewable and distributed power sources such as solar, wind, and hydrodynamic energy is decreasing.   These changes are altering the landscape for energy-intensive industries across the world. Amongst them: U.S. cable operators, who face a sea change in the profile of energy costs, as subscribers grow more hungry for video and bandwidth at the same time that the cost of power to support a typical cable operators’ plant is expected to grow steeply.

Cable operators have an opportunity to “get ahead of the curve” through structured decision analysis of choices about energy usage.  Quantellia’s Mark Zangari and Tim McElgunn, Chief Analyst of Bloomberg BNA Broadband advisory services, presented on this topic at the Society for Cable Telephone Engineers (SCTE) Smart Energy Management Initiative (SEMI) forum.  Through a combination of data provided by Bloomberg BNA and modeling provided by Quantellia, Zangari and McElgunn were able to demystify some of the important factors involving this complex decision making process.  McElgunn also authored an accompanying analysis report, which is available to Bloomberg BNA subscribers.

Says McElgunn, “Within the next five years, the cost of energy to power the U.S. cable industry will become the single largest network cost component.” As shown in the video below and described in more detail after the video, operators who make good decisions stand to benefit substantially.

https://youtu.be/hUiVk-ld3OE

Background – The Energy Cost of Success for Broadband Providers

In recent years, U.S. cable operators have steadily expanded their service offerings to both consumers and businesses in such a way that energy demand and associated costs will continue to increase.Bloomberg BNA research shows, for example, that over three quarters of U.S. internet service customers subscribe to “bundles” of two or more services, which may include pay TV, home phone, and wireless service in addition to high-speed data.

Cable operators are responding to this increasing demand by building new data centers and expanding existing ones.These data centers contain equipment with increasingly lower energy footprints – to as low as 3 watts per server blade.However, the total growth in energy demand exceeds the benefit from these technology improvements.

For these reasons, cable operators face difficult decisions in a complex environment.Where should facilities be located?How should existing local and/or regional infrastructure be leveraged?How should cooling and heating be planned?Should operators consider alternative energy sources to utility-delivered power?

Says  McElgunn, “The U.S. cable industry has effectively transitioned to an all-digital video plant, and is preparing for an eventual transition to an all-IP network.”A digital network is less costly to operate than an analog one, however, HDTV impacts this trend in important ways.Says McElgunn, “By the end of 2014, essentially all new television sales in the U.S. will be HDTV-compatible, increasing the number of HD-fed sets per home. Bloomberg BNA estimates that approximately 40 million U.S. cable households will subscribe to HD tiers in 2017, with 1.5 HDTVs connected per home. “A large number of these HDTV sets will be displaying HD linear channels, as well as video-on-demand (VOD).  A single HD video stream consumes approximately 8 Mbps, and there are multiple simultaneous such streams delivered to essentially every video household for some portion of each day.“This leads to massive demands on the network,” says McElgunn, “This will impact both distribution and access plant and the resources required for content generation and formatting.”

In addition, U.S. cable operators offer services to businesses, which place additional demands on networks, and the energy required to operate them.However, McElgunn and Zangari write that “Current architectures and software designs require that generation and transmission resources that are preparing and delivering all of that video operate at full power virtually 24 hours a day every day.”This applies to multiple parts of the cable plant: headends, data centers, and outside plant facilities, which together will require substantial growth in energy to power them.

Modeling a Solution

Quantellia worked with Bloomberg BNA to build a model that integrates forecasting data with a systems model to allow cable operators to quickly and precisely determine if a proposed energy usage strategy exposes the operator to unacceptable financial and/or technical risks, and to visualize how adjusting specific decision parameters impacts desired outcomes.

Among the variables included in the model are expansion of Network DVR, balancing increasing server efficiency versus equipment replacement costs, evolutionary network and facilities design changes, and energy price changes over time.

Some findings were as follows:

Capital costs are a substantial component: Each additional watt of power consumption adds incremental costs due to the increased quantity of energy the operator needs to supply.However, each additional watt also increases infrastructure needs, and the annualized capital costs resulting from higher power consumption can be significantly greater than the effect on power supply charges.

Moore’s Law effects are mitigated by power issues: As illustrated below, although the cost of delivering a given unit of computing performance has been dropping rapidly over the years, the cost of power required to deliver that performance has not been dropping at the same rate.

Timing is everything: Operators who take too long to change their energy contracts face steep competition through decreasing margins as costs increase.

Sleep mode versus peak shaving: Policies that invoke “sleep mode” for idle equipment can reduce energy usage, but it’s important to keep in mind that capital-intensive infrastructure must be built to support the “peaks” of usage, so “peak shaving” is also an important cost containment strategy.

Organizational Challenges

Despite the cost benefits from making good decisions about energy usage, says Quantellia’s Zangari, “most U.S. cable operators have separate organizations with independent budgets for the establishment and operation of facilities, and the operation of the network equipment housed in those facilities.  While power and its supply and assurance is managed as a facilities function, network equipment and its performance falls under the operational and financial responsibility of the Network organization.”  Therefore, there is no single department within a typical cable operator with a holistic view of energy management.

A model like the one discussed here, provides this kind of cross-organizational view and equips decision-makers with access to all the factors relevant to measuring, understanding, forecasting, and minimizing the cost of energy.

See http://www.bna.com/broadband-advisory-services-p12884902148/.

If you’d like a copy of this model to run yourself, please drop me a line.

Сижу готовлюсь к экзамену по менеджменту компьютерных ресурсов. Один из вопросов:"какова ваша модель принятия решений?" Моя модель принятия решений? Бросить все к чертям, когда все надоело... Быстро и безболезненно обрывать мосты. Купить бутылку вина и сыр с плесенью, когда в кармане 15 долларов. Забивать на все, когда у Даньки выдается свободная минутка. Просыпаться за 20 минут до пары и бежать сломя голову (при том, что будильник заведен на два часа до учебы). Всегда верить в лучшее. А если лучшее так и не наступает, то ехать к морю. Если и море не помогает, то устраивать идиототанцы)) они никогда не подводят. Делать то, что нравится, а если не нравится, то не делать, ибо без интузиазма я хрен когда что сделаю. Любить без памяти. Думать о будущем. Делать все в последний момент. Говорить то, что думаю, даже если это колко и не в попад. Не бояться упасть. Искать приключения на все мои точки. Мечтать много-много. Пялиться в воду или небо. Поступать по совести. Находить хорошее в людях и ценить их за это. Громко до слез смеяться, когда действительно весело. Непрофессиональный ответ на вопрос вышел... Конечно на экзамене я напишу все строгими и причесанными словами, но в голове так и останется моя спонтанная лохматая модель. Самое главное, при принятии решений, помнить, что у тебя только одна жизнь и шанс прожить ее так, чтобы потом не было мучительно больно от того, что что-то так и не сделал, так и не почувствовал, так и не сказал...