CASLab Logo Complexity Logo  

Research Projects

Innovation Ecosystem Modeling and Simulation

A novel approach for modeling innovation-based economic growth using computational agents and adaptive resource transformers.

Agent-based computational economics (ACE) has increased in popularity over the past few years as a tool to both understand and explain complex economic phenomenon. Econosim is a new ACE model being developed at the University of Central Florida. It differs from many existing ACE models by removing the distinction between firms and households.

In Econosim, every economic agent is both a consumer and a producer. This decision results in production and trade becoming the core economic behaviors of every economic agent. In addition, we explicitly represent production knowledge as a set of resource transformation rules that are subject to evolutionary forces. This representation allows population dynamics to alter the technological landscape of the economy and provides a straightforward method of exploring innovation and knowledge driven economic growth.  Each agent in our model, called an adaptive resource transformer (ART), lives within an economic ecosystem where individuals are connected by social networks and the actions of an agent can have unintended consequences beyond its nearest neighbors. Econosim is intended to serve as a computational economics laboratory that can be used to verify and explore existing economic ideas and theories, and help inspire and create new ones.  It is also intended to serve as a tool for exploring the impact of economic policy by allowing modelers to view the potential consequences of their decisions in silico, before they are enacted in the real world. The current model is written in Java, using the MASON toolkit. It is being employed to explore the benefits of entrepreneurial support organizations, such as university incubators, and examine the structure and influence of economic networks.

Business Incubation & Acceleration Dynamics

Modeling the role of business incubators considering different support methods and show their effects on global economy with our preliminary results:

The innovation ecosystems are clusters of economic entities creating high productivity and business diversity based on innovations. The business incubators are among the most well-known approaches to support economic growth in these ecosystems. Although the number of business incubators has increased substantially in the last decade, the absence of efficient assessment techniques for business incubation calls for novel approaches. We model the role of business incubators considering different support methods and show their effects on global economy with our preliminary results.

Incubation Impact Analysis

Modeling the role of business incubators considering different support methods and show their effects on global economy with our preliminary results:

Business incubators have become a popular policy option and economic development intervention tool. However, recent research shows that incubated firms may not benefit significantly from their incubator relationships, and may even be more vulnerable to failure post departure (graduation) from an incubator. These findings suggest that the impact of business incubation on new venture viability may be contingent on the type of support offered by an incubator and attributes of business environments within which incubation services are provided. Incubation services that protect and isolate ventures from key resource dependencies may hinder venture development and increase subsequent vulnerability to environmental demands. Alternatively, incubation services that help ventures connect and align with key resource dependencies are likely to promote firm survival. We propose that incubators vary in the services and resources they offer, and that university incubators typically provide greater connectivity and legitimacy with respect to important contingencies associated with key industry and community stakeholders. This leads us to propose that university affiliation is an important contingency that affects the relationship between firms’ participation in incubators and their subsequent performance.

The purpose of this study is to evaluate this contingency by examining whether firms graduating from university incubators attain higher levels of post-incubation performance than firms participating in non-university affiliated incubators. We test this by evaluating the performance of a sample of graduated firms associated with the population of university-based incubators in the US contrasted against the performance of a matched cohort of non-incubated firms. The analysis uses an enhanced dataset that tracks the number of employees, sales, and the entry and graduation (departure) points of incubated firms from a university incubation program, so as to delineate the scope of influence of the incubator.

Data Analytics for Higher Education

Improving STEM Education Using a Participatory, Social and Collaborative Architecture for Learning:

This research focuses on understanding the most effective ways to use information technologies for STEM undergraduate education. We have partnered with the Mathematics department and focus on Calculus I and II classes. The Discuzz project is a partnership with Microsoft Corporation to use their emerging social networking technologies (Discuzz system) in a calculus II class the Summer 2012 and use the other five Calculus II sections as control group.

Modularity in Genetic Algorithms

Investigating the question of what constitutes a module at the genomic level of evolutionary search:

Modularity is a common characteristic of many artificial and natural systems. Systems with separable or nearly separable units are considered modular. Alternatively, modules may also viewed as repetitively used design units [21] that are easy to dissociate, recombine, and reuse in different systems. In this study, a module is a sequence of genomic primitives that may be repeated in the genome or reused throughout the population. A genome includes actual system primitives as well as lower level encapsulated module names. This way of defining modules relates to Holland's building blocks hypothesis. Modules, however, differ from building blocks in their robustness against recombination operators. An encapsulated module cannot be disrupted by recombination while building blocks can.

In this study, we prove, under a set of assumptions, that the systematic encapsulation of lower order modules into higher order modules does not change the size or bias of a search space and that this process produces a hierarchy of equivalent search spaces. We also investigate the question of what constitutes a module at the genomic level of evolutionary search and provide a static analysis of how to identify good and bad modules based on their ability to reduce the search space, thus, biasing the search space towards a solution.

Curriculum GPS: An Adaptive Curriculum Generation and Planning System

Growing and maintaining programs with high retention and satisfaction rates in college, military or corporate education:

In educational systems, there has been an increasing interest for the innovative applications such as educational data mining and predictive analytics. These applications are utilized by the institutions for fulfilling academic missions and for improving the utilization of institutional resources. In this paper, we propose the “Curriculum GPS”, an adaptive curriculum generation and planning system, to provide a quantitative model and an interactive system that helps to grow and maintain programs with high retention and satisfaction rates in college, military or corporate education.

The Curriculum GPS is composed of three main components: Curriculum analysis, historical data mining and an adaptive course sequence generation. The existing literature demonstrates how curricular efficiency correlates to student academic success in terms of graduation and retention rates. Therefore we first use an approach from the literature to analyze the curriculum under discussion as a directed graph by considering the conditions among courses such as prerequisite requirements. We conduct network analysis in this graph and compare our results with the catalog of courses currently in use. Then we combine this analysis with the historical data of the students and courses to train our model and develop our system’s database. The resulting system uses this training to create a set of quantitative recommendations for each student depending on her individual data such as passed/remaining courses, grades and time to graduate. The system also allows running what-if scenarios to test the outcomes of different choices by students. Therefore it is advantageous for students, instructors and advisors. The system is being developed for the Information Technology based departments of one of the largest universities in US by using the curricula and student datasets from the last thirty semesters. Initial results suggest this novel system provides both insight and improvement for the institutional education.

Innovation Ecosystem of Optics Photonics Industry

Identification of the impact of research activities on the regional entrepreneurship for a specific high-tech industry:

The economic entities in innovation ecosystems form various industry clusters, in which they compete and cooperate to survive and grow. Within a successful and stable industry cluster, the entities acquire different roles that complement each other in the system. The universities and research centers have been accepted to have a critical role in these systems for the creation and development of innovations. However, the real effect of research institutions on regional economic growth is difficult to assess. In this study, we present our approach for the identification of the impact of research activities on the regional entrepreneurship for a specific high-tech industry: optics and photonics. The optics and photonics has been defined as an enabling industry, which combines the high-tech photonics technology with the developing optics industry. The recent literature suggests that the growth of optics and photonics firms depends on three important factors: the embedded regional specializations in the labor market, the research and development infrastructure, and a dynamic small firm network capable of absorbing new technologies, products and processes. Therefore, the role of each factor and the dynamics among them must be understood to identify the requirements of the entrepreneurship activities in optics and photonics industry.

The recent studies show that the innovation in optics and photonics industry is mostly located around metropolitan areas. There are also studies mentioning the importance of research center locations and universities in the regional development of optics and photonics industry. These studies are mostly limited with the number of patents received within a short period of time or some limited survey results. Therefore the first contribution of our approach is conducting a comprehensive analysis for the state and recent history of the photonics and optics research in the US. For this purpose, both the research centers specialized in optics and photonics and the related research groups in various departments of institutions are identified and a geographical study of their locations is presented. The second contribution of the paper is the analysis of regional entrepreneurship activities in optics and photonics in recent years. We use the membership data of the International Society for Optics and Photonics (SPIE) and the regional photonics clusters to identify the optics and photonics companies in the US. Then the profiles and activities of these companies are gathered by extracting and integrating the related data from the National Establishment Time Series (NETS) database, ES-202 database and the patent database. Our third contribution is the utilization of collected data to investigate the impact of research institutions on the regional optics and photonics industry growth and entrepreneurship. In this analysis, the regional and periodical conditions of the overall market are taken into consideration while discovering and quantifying the statistical correlations.




UCF Complex Adaptive Systems Laboratory
12800 Pegasus Drive, Lab 314
Orlando, FL 32826-3246
Phone: 407.882.1163
Ivan Garibay, Director