Estimate Consumer Load in 3G Network Migration

wifi_coverage


Challenge
Migrating from a 2G network to a 3G network, Orange considered carrying 20 % of phone calls over WiFi to decrease the load of 3G cells and save on deployment costs.

Approach
Icosystem identified the critical mass of WiFi coverage needed to transfer a significant capacity. Sensitivity analysis showed that WiFi coverage is strongly influenced by WiFi density and also by WiFi sharing.

Outcome
We can use mobility data to analyze any type of consumer behavior that is dependent on daily fluctuations.
An optimizer can find the most profitable business strategy.

 

Sample Case Study


Challenge

Approach

Outcome

 

Earn Better ROI Understanding Consumer Decision-Making

“Icosystem’s contributions to understanding and predicting human decisions are always fresh, innovative, surprising and have helped us tackle some of the most difficult problems of marketing and media planning.”

– Masakazu Okano, Deputy Director, Marketing and Media Planning Division

With its focus on consumer-centric accountability and cross-media ROI, the tool has helped Dentsu earn not just more work but also satisfaction from its clients.

 

Driving Traffic with Pricing & Advertising

GSD&M IdeaCity is one of Omnicom Group’s largest advertising agencies. Using agent-based (consumer-centric) modeling to model the casual dining ecosystem for a GSD&M client, Icosystem was able to make traffic predictions based on pricing and media plan within 5% of actual numbers, making the tool the most accurate predictor in the space. According to Maury Giles, VP, Analytics and Accountability, Icosystem’s technology can help save agencies and their clients millions of dollars.

 

Accurate Prediction of Malware Intervention Strategies

“We’re very pleased with the results of this work.”

–Paul Waldron, Director of Fraud Operations, France Telecom

“We worked with Icosystem to develop a software model that allows us to explore the development of an outbreak of mobile malware in a population of mobile phones. Icosystem’s model was able to predict the future evolution of the infection with enough accuracy to give us confidence that we could use it to investigate intervention strategies that could be used to slow or control the outbreak. Icosystem’s research staff were extremely flexible in their approach to the project and proactively brought forward ideas on how we might address the challenges in building this kind of model.”

 

Test Impact of Personnel Policies Using Flexible Modeling Platform

ss_case_study_personnel_STEER1


Challenge
The Navy’s FUTURE program required a flexible modeling platform for its Manpower, Personnel, Training & Education (MPT&E) needs to test different types of Navy organizational structures and compare them.

Approach
Icosystem developed the Simulation Toolset for Experimental Environment Research (STEER), a simulation environment platform to support MPT&E decision makers. The STEER toolset provides a platform for developing and testing models within the MPT&E domain. STEER is not a specific model but rather an environment that provides a flexible, extensible core that can be customized towards a specific class of end-user domain-centric models. For example, STEER can be used to develop a simulation tool with a highly-customized user interface and analytics/reporting capabilities for Strength Planners or for Community Managers. This allows for successively more powerful models to be developed upon a trusted/tested simulation platform.

[More]

Accurately Forecast Personnel Staffing

ss_case_study_personnel_IMPACT1
ss_case_study_personnel_IMPACT2

ss_case_study_personnel_IMPACT3


Challenge
To maintain a reserve force with the proper level of training and expertise, personnel planners must accurately forecast overmanning, undermanning, and other risks to guide accessions, promotions, training, and personnel policy. Personnel inventory is influenced by a complex interaction of external factors (e.g., local and national economic conditions, civilian/government compensation differences) and internal ones (e.g., advancement policy, participation requirements, bonus offerings), making it very difficult to forecast changes using traditional statistical approaches.

[More]

“Co-Evolving Business Models: A Case Study with the Internet Service Provider (ISP) Industry”, New Mathematics and Natural Computation, 2005.

Order online through the NMNC website or request a free reprint from Icosystem.

In this paper, co-evolution is used to examine the long-term evolution of business models in an industry. Two types of co-evolution are used: synchronous, whereby the entire population of business models is replaced with a new population at each generation, and asynchronous, whereby only one individual is replaced.

 

“Model Behavior: The Way a Company Really Works is Probably Not the Way Managers Think It Does”, Optimize, 2002.

Download (PDF)

Managers don’t need pictures of hierarchy; they need visualizations of the wide-ranging connections that make up companies’ learning systems. Rather than charts showing who reports to whom, they need charts to show who knows what and whom, and who works most often with whom. That’s the purpose of organizational-network analysis (ONA), the application of social-network theory to organizations. ONA paints a much more accurate picture of how a company actually works, shares knowledge, and completes processes.

[More]

“A Drug Candidate Design Environment Using Evolutionary Computation”, IEEE Transactions on Evolutionary Computation, 2008.

Download (PDF)

This article describes the Candidate Design Environment we developed for efficient identification of promising drug candidates. Developing effective drugs from active molecules is a challenging problem which requires the simultaneous satisfaction of many factors. Traditionally the drug discovery process is conducted by medicinal chemists whose vital expertise is not readily quantifiable. Recently, in silico modeling and virtual screening have been emerging as valuable tools despite their mixed results early on. Our approach combines the capabilities of computational models with human knowledge using a Genetic Algorithm and Interactive Evolutionary Computation. We enable the chemist’s expertise to play a key role in every stage of the discovery process. Our evolved structures are guaranteed to be within the chemistry space specified by the medicinal chemist, thereby making the results plausible. In this paper we describe our approach, introduce a case study to test our methodology, and present our results.

 
« Newer items — Older items »

index.php

Icosystem Corporation • 10 Fawcett St., Cambridge, MA 02138 • Voice: (617) 520 1000 • Fax: (617) 492 1505

Terms of Use • Privacy Policy