Shopping Cart
Your Cart is Empty
Quantity:
Subtotal
Taxes
Shipping
Total
There was an error with PayPalClick here to try again
CelebrateThank you for your business!You should be receiving an order confirmation from Paypal shortly.Exit Shopping Cart

RWISE Academic Publications and White Papers

Reference World offers many FREE downloadable products, charts and papers. Feel free to browse and register for any products you are interested in.


Each link will take you to a registration page that you will need to fill out to receive our products. We use this information to track our audience and do not share this information externally with third parties.

By filling out our form, you agree to receive email communications from Reference World containing products and information you register for.

Addressing the Challenges of Data-Driven Analysis in Intuition-Driven Organizations

-2020-

C. F. Day

Charles F. Day & Associates

Fredericksburg, VA

B. Armstrong

Charles F. Day & Associates

West Lafayette, IN

Summary

Over the last 15 years, management’s adoption of technology has advanced much slower than technology itself. Primary challenges include holding informed discussions with senior management, shaping realistic understanding of capabilities, working with stove piped information technology departments with a restricted understanding of new technology, and acceptance of data-driven simulation results over intuition-based expectations. These challenges grow in magnitude with organizations driven by intuition-based decision makers.


Biases and knowledge gaps with modeling and simulation (M&S), artificial intelligence (AI), and machine learning (ML) among managers and leaders are evident across a spectrum of clientele. When organizations decide to move from intuition-based to data-driven decision, they encounter hurdles of biases, gaps, poor assumptions, and emotion-driven responses.


In this paper, we describe the value of tying the learning life cycle directly to the business development process from lessons learned from our own experiences. Many challenges can be addressed through organizational development theory, tools, and techniques. Others can be clarified by examination of the cognitive bias codex and personality preferences. Integrating the learning and business development cycles may be the best way to overcome the technology disillusionment curve.


Our experience dates to Synthetic Environment for Analysis and Simulations (SEAS), an agent-based M&S platform and NTSA 2004/2005 award winner in analysis. We’ve moved forward to Reference World Synthetic Information Environment (RWISE) an agent-based M&S platform enriched with AI/ML in an elastic environment. SEAS was years ahead of its time with distributed AI operating in a hybrid cloud. It stretched the imagination and challenged intuition-based decision-making in the Department of Defense, Homeland Security, and commercial users in broad range of areas. Its successor, RWISE, is a data agnostic, data driven, agent-based AI/ML platform for data ingestion, model development, and forecasting to compare multiple futures based on injecting and testing strategy actions.

Register to Receive Download

Forecasting the Impact of Policy Interventions on Societal Behavior

-2020-

B. Armstrong

Charles F. Day & Associates

West Lafayette, IN

C. F. Day

Charles F. Day & Associates

Fredericksburg, VA

A. R. Chaturvedi

Purdue University

West Lafayette, IN

Summary

To overcome the challenge of heterogeneity in data sources, format, and content, disciplinary theories, and modeling frameworks we describe a methodology that partitions all data into individuals, organizations, institutions, infrastructure, and geographies entities (IOIIGs) and automatically extracts relevant theories for each entity class. We then reconstruct a virtual Reference World (RW) in which IOIIG “agents” interact with each other to produce emergent societal level behavior. Policies are defined in terms of actions by some Institution and Organization agents to shape the behaviors of Individual agents so that desired societal level outcome may be achieved. The RW approach consists of four phases: simultaneous self-ingestion of data with continuous verification and validation, self-extraction of behaviors, self-extraction of policies, and simulation of future states.


RW methodology continuously ingests dynamic data through an onboarding process which relies on semantic definitions of fields. Static traits and dynamic actions of IOIIGs are extracted as sampled data to form behavior maps for the various modeled entities. Data describing intended and observed impacts of policy actions are converted into action meshes. The behavior maps and action meshes are then combined using neural network and machine learning techniques to create plausible future meshes. Agent based simulation is then used to produce future states.


We demonstrate the application of the RW approach to supporting strategic policy design for increasing the postsecondary attainment of the work eligible population nation-wide.

Register to Receive Download

Securing the Food Supply Chain: Understanding Complex Interdependence Through Agent-Based Simulation

-2014-

A. R. Chaturvedi

Purdue University

West Lafayette, IN

B. Armstrong, R. Chaturvedi

Charles F. Day & Associates

West Lafayette, IN

Summary

The food industry has many points of vulnerability in its supply chain. It currently lacks integrated crisis management and response programs to understand the importance of decision-making during and in the aftermath of a bioterrorist attack on the food supply. Computer simulations have been used successfully in other industries as training and analysis tools. This paper describes an agent-based simulation for food defense training and analysis. Production information, consumption patterns, morbidity/mortality rates, recall costs and additional information were collected and provided to a data driven simulation to anticipate the impact of decision-making on economic and public health during a terrorist attack. A case study is given with a representative exercise involving forty industry representatives who participated in a food defense simulation. Their decisions (recall and microbiological and toxicological testing) were derived from testing results, press releases, epidemiological data, and discussions with other industry and regulatory teams. Decisions made during the simulation resulted in over 76,000 illnesses, 45 deaths, and $132 million in recall costs. The no intervention, baseline scenario estimated to result in 91,000 illnesses and 54 deaths, indicating the improved public health outcomes resulting from players’ decisions. Participants identified three key learning points:

  

   1) Communication between all groups is pertinent and challenging.
   2) Approaches to solve inherent food safety problems cannot be used to address food defense situations.
   3) Human resource procedures regarding new hires and disgruntled employees should involve additional security measures.


This computer simulation could be a valuable resource in food defense awareness and help educate companies and regulators about food defense risks and decision-making consequences.

Register to Receive Download

Got a Problem? Agent-Based Modeling Becomes Mainstream

-2013-

R. Chaturvedi, B. Armstrong

Charles F. Day & Associates

West Lafayette, IN

A. R. Chaturvedi

Purdue University

West Lafayette, IN

D. Dolk

Naval Post Graduate School

Monterey, CA

P. Drnevich

University of Alabama

Tuscaloosa, AL

Summary

Agent-based modeling and simulation (ABS) is emerging as a key technology that is helping to enhance the understanding of social sciences. Systems ranging from organizations to economies and societies can be modeled to provide insights in ways that were previously not possible with quantitative approaches. The Sentient World Simulation (SWS) is an ultra-large-scale ABS developed to capture a comprehensive view of “Whole of Government” operations. The SWS supports a strategic geopolitical perspective that captures the interplay between military operations and the social, political, and economic landscapes. The SWS consists of a synthetic environment that mirrors the real world in all its key aspects. Models of individuals within the synthetic world represent the traits and mimic the behaviors of their real-world counterparts. As models influence each other and the shared synthetic environment, behaviors and trends emerge in the synthetic world as they do in the real world. The SWS reacts to actual events and incorporates newly sensed data from the real world into the virtual environment. Trends in the synthetic world can be analyzed to validate alternate worldviews. The SWS provides an open, unbiased environment in which to implement diverse models. This results in a single holistic framework that integrates existing theories, paradigms, and courses of action.

Register to Receive Download

What people are saying about us

The world is not flat. Neither should a world model be flat.

We use agent-based modeling to create a multi-dimensional Reference World.

Alok Chaturvedi - Purdue University