|
A Note from The Analytics Lens Newsletter Editor
Michael F. Gorman, University of Dayton
I am pleased to present our second newsletter. Thanks to all who contributed! In this newsletter we discuss the INFORMS Annual Meeting analytics track and other analytics sessions of interest to our members, results of a recent survey on analytics, ruminations on analytics now and then, and another analytics story. I hope you enjoy reading it.
We look forward to our winter newsletter. Feel free to suggest content! Contributors are needed and welcome. Please let me know if you have some insights to share. Thanks for your support!
|
|
A Letter from the Analytics Section President
Dear fellow Analytics Section member, The Analytics Section continues to grow and develop at a rapid pace. I have a number of developments to report.
- Great Deals - INFORMS has approved a $30 discount for the spring INFORMS Analytics Conference for all Analytics Section Members.
- Free Offers to Members - we have been given free passes to IE Analytics Conference in Chicago and to Aberdeen Analytics Summit in Boston this year with a promise of more free passes to come.
- Exciting News - In this newsletter, we announce an "Innovations in Analytics" award that will be sponsored by our Section. This is an exciting development for us and all of INFORMS. More information below.
- Informative Sessions - We have co-sponsored a 13-session Analytics Track in the INFORMS Annual Meeting - see track description below. Attend the conference!
- Stay in Touch - The Analytics Section has an active discussion board on LinkedIn where numerous jobs have been posted, and an active discussion on "What is Analytics", among others, is going on.
- Great Momentum - As a result of our activities, and the inherent interest in analytics, our membership has grown to over 450 people, which makes it the fastest growing new sections in INFORMS.
- Next Up - Plans are in the works to have an active presence in the spring 2012 INFORMS Analytics Conference. Feel free to help us plan.
As we grow, we need your help. Remember - we are looking for bloggers for the website, tweeters at the conference, and we are actively looking for a webmaster for our section's website. Let me know if you are interested! I look forward to seeing you all at the fall INFORMS Annual Meeting. Sincerely, Michael F. Gorman INFORMS Analytics Section President University of Dayton Department of MIS, OM and DSC President, MFG Consulting, Inc.
|
|
Why Renew Your Analytics Section Membership? You've Seen the Reasons.
Michael F. Gorman
You've seen our activities - not bad for our first half year! We know you share our excitement and enthusiasm for analytics, which truly makes a difference in the lives of people everywhere. We invite you to renew your membership in the section today. Your membership dues show your support for our vitally important analytics field and guarantee uninterrupted access to exclusive member benefits.
- Discounts to the spring INFORMS Analytics Conference, opportunities for free admission to other conferences
- On our blog, LinkedIn group, and at our conferences, you can forge new relationships and network
- You will have access to inside insights and information to state-of-the-art analytics information sharing through our newsletter and conferences
- Present, and gain recognition for your work
- Gain access to Section postings, find hidden jobs and advance your career
- We will extend our service to professionals in the analytics field to help develop analytics capabilities in organizations
- We will help to shape and understand analytics curricula at universities
Through all of this, we will continue to grow our numbers and influence on the INFORMS organization. With the fall Annual Meeting comes INFORMS renewal notices. Above are some reasons why you should renew your Analytics Section Membership for just $20 for INFORMS members, $50 for non-INFORMS members, and $5 for retired and student members. Your membership dues support all of these activities, plus much more, including educational programs and outreach activities to spread the good word about analytics. Please don't delay-renew your Analytics Section of INFORMS membership today. Save Time - Renew Online! https://renew.informs.org
|
|
2012 Innovation in Analytics Award
Michael F. Gorman, Innovation in Analytics Award Committee Chair
I am pleased to announce the "Innovation in Analytics Award," an award created and sponsored by the Analytics Section of INFORMS! The purpose of the Innovation in Analytics Award is to recognize creative and unique developments, applications or combinations of analytical techniques. The prize promotes the awareness of the value of analytics techniques in unusual applications, or in creative combination to provide unique insights and/or business value. The prize is not meant to recognize strictly theoretical advances, though theoretical advances might be an enabler to innovative applications. Similarly, the prize is not focused on implementation value created, but such value creation might add credibility to the innovation. Applicants can submit proposals in the descriptive, predictive or prescriptive areas of analytics. Descriptive innovation could include novel measures, visualization methods, or creative approaches to the analytics process improvement among others. Predictive innovation might be statistical application, text analytics or simulation, among others. Prescriptive innovation might include, for example, game theory or optimization. The lists above are not meant to be exclusive, but rather illustrative examples of areas of innovation. The application process is as follows:
- Applicants should submit a brief (500-1000 word) summary of their work by January 31, 2012.
- Semifinalist applicants may be queried by a subset of the committee to clarify, verify and validate their work via either written or oral questions (email or phone call) during the month of February.
- Finalist(s) will be selected based on their summary, presentation and response to judge's questions by March 1, 2012.
- Finalist will present their work in a closed-door, judges only, session at the spring INFORMS Analytics Conference.
- The winner will present their work in an open 45-minute session at the conference.
The judging criteria will be based on the level of uniqueness, creativity, value and (potential) contribution of the analytical techniques developed. Submissions will be evaluated based on innovations in any of the three dimensions of analytics: descriptive, predictive, or prescriptive. Submissions that span more than one dimension of analytics are preferred. Implemented approaches are preferred but not required; to the extent implemented savings are claimed in the submission, the judging panel will seek verification of the implementation. Submissions that have the potential for more widespread use will be given added consideration, but particularly clever innovations with narrow focus are also highly valued. The prize is intended to recognize recent or on-going work. Any member of INFORMS is eligible to submit an application. In order to win the Innovation in Analytics award, the applicant must be willing to share the innovation in a public presentation or publication, as described elsewhere in this document. No Analytics Section officer or Innovation in Analytics Award judge is eligible. The prizes for the finalists are as follows: winner $1000, second place $500, third place $250. Each of the finalists will receive a plaque commemorating their accomplishment. Finalist(s) are required to present at the spring Analytics Conference. The winner will have a dedicated session at the spring conference. All finalist(s) will be expected to write a summary of their work for Analytics Magazine, OR/MS Today, Interfaces, and/or The Analytics Lens. The award committee consists of: Michael F. Gorman, University of Dayton, Committee Chair; Gavin DeNyse, Hewlett Packard; Pooja Dewan, BNSF Railway; Juan Jaramillo, Albany State University; Patricia Neri, Chase; and Vijayakumar Ramdoss, IBM. For more information, visit our website. All submissions should be sent to [email protected].
|
|
INFORMS Annual Meeting Analytics Track
Manoj Chari, Practice Chair, INFORMS Charlotte Meeting and Co-Chair, Analytics Cluster
The Analytics Section is sponsoring a newly created Analytics Cluster at the INFORMS Annual Meeting and on behalf of my fellow Co-Chairs Mike Gorman and Doug Samuelson, I would like to invite you to actively support this new cluster. Please plan to attend the sessions in the cluster and encourage other INFORMS colleagues, especially practitioners, to attend as well. As always, feedback about the section in this conference, as well as ideas about and active participation in future conferences is welcome. Please note that for the Charlotte conference the Analytics Cluster subsumes the Practice Cluster, and is also co-sponsored by CPMS. The Analytics Cluster in the Charlotte meeting has a rich variety of presentations and panel discussions that cover a wide range of topics that can be broadly be classified into two categories:
- The Analytics Discipline and its Practice - To familiarize the INFORMS community with the discipline, the cluster includes several sessions that delve into the nature of the analytics as well as important aspects of analytics practice and the analytics process. For those members who are new to the analytics initiative, there will be an interesting session on the genesis of the analytics movement within INFORMS and its possible future prospects led by INFORMS Vice President for Practice, Jack Levis and Mike Gorman, who is also the President of the INFORMS Analytics Section. There will also be panel discussions and presentations on topics such as the assessing the analytical maturity and opportunity in an organization, the selling of the value of analytics within an organization, the interface between analytics and IT - all of which are key elements of successful implementations of analytics in business. Two sessions on analytics education - one each covering undergraduate and graduate programs- will address the critical issue of preparing future analytics professionals. There will be also be tutorial/overview type presentation sessions from two major analytics vendors SAS and IBM on predictive analytics and descriptive analytics which are two areas that complement the typical INFORMS focus on prescriptive analytics.
- Specific Applications of Analytics - Leading this category of talks on success stories of analytical applications are reprise sessions of the 2011 Edelman finalists. These represent the most impressive achievements of operations research and analytics practice in the past year as judged by INFORMS. We will also have a session devoted to applications in companies that are members of the INFORMS Roundtable. In addition, leading industry organizations will have speakers participating in sessions devoted to applications in specific verticals, and these include public sector, petrochemicals, energy, finance, and security/law enforcement.
I look forward to seeing you all in best wishes for a successful meeting in Charlotte.
|
|
Fall Business Meeting Announcement
Please attend the Analytics Section Business Meeting on Monday, November 14 at the Convention Center, Room 203A, from 6:15 to 7:15pm.
Thanks to SAS for sponsoring the food and drinks! |
|
Are You Going to Charlotte? Let Us Know - Quick Survey
We are planning for our business meeting and other sessions for Charlotte, and would like to know if you are coming. Early responses indicate that 65% of Analytics Section members will be attending, almost all will be attending at least some Analytics Track sessions, and almost half will be attending the business meeting! Help us plan in just five clicks.
If you haven't already, would you please take the time to take this three question survey?
|
|
Other INFORMS Conference Sessions of Interest to Analytics Members
There are a number of sessions at the upcoming INFORMS Annual Meeting that are not in the Analytics Track, but could be of high interest to analytics professionals. Here are some suggestions.
A quick search of the program reveals almost 100 presentations with the keyword analytics in it! Lots to choose from - but which ones?
The keynote on Monday is surely to be of interest, "The Evolution of Analytics" will be presented by Keith Collins, Senior Vice President and Chief Technology Officer of SAS.
Disciplines like operations research, statistics, forecasting, and econometrics have distinct roots and continue to develop in their own right - in research and academia. In practice it's increasingly recognized that they have grown together, producing a hybrid known as "business analytics." What is propelling the business analytics trend, what problems does it seek to address, and where is it going? Come listen! Other interesting subjects:
- A track on sports analytics - baseball revenues, (fantasy) football predictions, and individual player contribution
- Analytics in industries such as rail, security, finance and others
- Major companies like UPS, Sasol, and Chevron will share their analytics experience and successes
- Major software suppliers like SAS and IBM will share their insights on analytics software developments and options
- Vendor specific training on analytics software
- And much more.
Please spelunk the program at your leisure to see which of your topics of interest might be covered! Follow the link below and click on "Search the Program".
http://meetings.informs.org/charlotte2011
|
|
What is Analytics? Part 2
A major task and opportunity for Analytics Section members is to help define analytics and its role in INFORMS. This has been a popular topic on our discussion board.
In his book, Competing on Analytics, Tom Davenport defines analytics as:
"The extensive use of data, statistical and quantitative analysis, explanatory and predictive models, and fact-based management to drive decisions and actions. Analytics are a subset of what is now known as business intelligence."
Matthew Liberatore and Wenhong Luo, in their 2009 Interfaces article add to Davenport's definition that in general, "analytics is the science of logical analysis", and more specifically is "a process of transforming data into actions through analysis and insights in the context of organizational decision making and problem solving".
In the last newsletter, we started a conversation called "What is Analytics, Anyway?" and posted a discussion on the LinkedIn site.
Here is what we have heard in response:
Irv Lustig
Personally, I'd agree with the first sentence (of Davenport's), and disagree with the second sentence. I see BI as a subset of analytics. And O.R. is one piece of the analytics pie.
Mike Gorman
Analytics is about creating a vision, selling an idea, seizing the opportunity, selecting of tools and technologies, managing the implementation process, mapping the data requirements and systems integration, and leading and sustaining the organizational change that allow O.R. to deliver on its promise.
Zahir Balaporia
I believe analytics is essentially about how we make decisions, rather than the decisions we make. And regardless of the complexity of the analytical models we employ to inform our decision making, the most important model is the one that sits between our ears i.e. our mental model. So analytics should be about improving our mental models/understanding/insight about a problem so we can make better decisions.
Christopher Berry
The application of the scientific method to data for the purpose of deriving sustainable competitive advantage or solving a societal problem.
Bill VanMarter
'New' analytics pays conscious attention to business questions and process as well as technology, then the 'old' Business Analysis in the guise of traditional O.R. focuses primarily on improved/optimized technology within an existing process. Traditional BA and OR tend to remain in the existing technological box. Analytics moves out of this box into a larger one including the process reexamination.
What do you think? Add your comments to the LinkedIn discussion today! We plan to discuss more at our sessions in Charlotte!
|
|
Summary of Relationships Between O.R. and the Analytics Survey Results
Matthew J. Liberatore, PhD; Wenhong Luo, PhD
The purpose of this survey research was to investigate the perceived differences between O.R. and analytics and their implications for O.R. professionals in terms of 1) data requirements, analytical methods, and insights; and 2) needed quantitative and soft skills. A total of 8,335 emails were sent to INFORMS members with 8,217 delivered, and 3,074 were sent to Analytics magazine readers with 2,996 delivered, for a total of 11,409 sent and 11,123 delivered. Since some Analytics magazine readers are INFORMS members, 566 individuals received both emails, with 558 delivered. Therefore, the total sample size was 10,565. A total of 2,245 responses were received and 2,195 surveys were retained, for a response rate of 2,195/10,565=20.8%. We limit the respondents to those who are familiar (4 or 5 on a 1 - 5 Likert scale) with analytics and O.R., with a total of 1,206 usable responses obtained. The vast majority of the respondents are fairly evenly distributed between 25 and 64 (91%), male (83%), and from the US/Canada (79%). Nearly all (96%) possess an advanced degree, and half (50%) are practitioners.
The mean values for the three data requirements items (availability, quantity, quality) were significantly higher for analytics (4.64, 3.99, 4.45) than O.R. (3.87, 3.30, and 4.07). This might be expected, since analytics starts with data while O.R. is more model-driven. However, data issues are still very important to O.R., as evidenced by the high mean scores. In terms of analytical methods, the mean values for descriptive and predictive methods are significantly higher for analytics (4.50 and 4.57 vs. 3.89 and 3.70), while the mean value for prescriptive methods is significantly higher for O.R. (4.79 vs. 3.75). Again, these results confirm our expectations, but it should be remembered that all three categories of methods were found to be important to both analytics and O.R. Along the same lines, in terms of insights provided, the mean values of "what happened" and "what will happen" are higher for analytics than O.R. ( 4.48 and 4.25 vs. 3.33 and 3.91), while the mean value for "what can happen" is higher for O.R. (4.72 vs. 3.81).
We compare the skill requirements for analytics and O.R. in two ways. First, we rank and compare the most important skills for analytics and O.R. Second, we assess each skill to see if it is perceived as more or less important for O.R. or analytics. The top five quantitative skills for analytics are: statistics, data visualization, data management, data mining, and decision analysis. The top five for O.R. are quite different: optimization, decision analysis, simulation, programming, and risk analysis. On the other hand, the same set of soft skills are ranked within the top five for both analytics and O.R. However, the order of ranking for analytics is data presentation, business knowledge, communication, problem recognition, and problem formulation, while for O.R. the last two skills become the top ranked ones. The results in toto emphasize the importance of a broad range of both quantitative and soft skills for O.R. and analytics, since all skills had mean values greater than 3.00.
Concerning quantitative skills, data management, data mining, data visualization, and statistics were significantly higher for analytics. These results reflect the stronger emphasis on data issues and descriptive and predictive modeling for analytics previously discussed. The values for decision analysis, optimization, programming, and simulation were significantly higher for O.R. This is in line with expectations, since all of these skills are generally acknowledged to be important for O.R. professionals. Interestingly, no significant difference between analytics and O.R. was found for risk analysis.
A set of business and consulting-oriented soft skills were found to be higher for analytics than O.R.: business knowledge, change management, communication, data presentation, interpersonal, metrics/KPI determination, persuasion, and project management. These results may reflect the view that analytics is an important business function, requiring stronger business skills. On the other hand, skills traditionally associated with O.R., namely problem formulation and problem recognition, were found to be higher for O.R.
Surprisingly, the respondents perceive that both fields are more closely related than might be expected. A key distinction between analytics and O.R. is the role of data in the process. For some the heavy emphasis on prescriptive methods is seen as the differentiator between O.R. and other fields such as analytics. Concerning the soft skills results, a possible explanation is that analytics is viewed as a business function, so analytics professionals are perceived to require more business-oriented soft skills.
The relationship between O.R. and analytics can be thought of as an intersection, but we believe that this intersection is not limited to "advanced analytics." The intersection cuts across not only methods used, but also quantitative and soft skills employed. Although both fields emphasize different methods and skills, a common set is important to both fields. At the risk of overgeneralization, analytics addresses improved decision making by gathering relevant, clean data first, followed by analysis, while O.R. focuses on developing and applying relevant models and then gathering the data to support them. Both approaches make progress toward the goal, and each can inform the other to improve. To better and more fully participate in the analytics revolution, O.R. needs to placed added emphasis on data management and analysis skills, and apply the soft, business-oriented skills so necessary for successful project and change management.
|
|
Thank you to SAS for sponsoring the food and drinks for the Analytics Section Business Meeting at the 2011 INFORMS Annual Meeting!
|
|
