Semen Koksal

Emeritus Faculty | College of Engineering and Science

Contact Information
(321) 674-7714
Frederick C. Crawford Bldg, 308D

Personal Overview

UBM-Group: Research and Education Program in BioMath (REP) funded by NSF. Duration: 36 months starting at Jan.1, 2008.

Educational Background

B.S. Middle East Technical University, Turkey
M.S. Middle East Technical University, Turkey
Ph.D. Florida Institute of Technology 1992

Professional Experience

Dr. Koksal’s areas of expertise are stability analysis of nonlinear ordinary differential equations, neural networks and mathematical modeling.

Selected Publications

1. Cowan J. and Koksal S. Fuzzy Backprobagation Networks using Vector Valued Activation Function, IEEE, NN Transc. (submitted).

2. van Woesik R and Koksal S A Coral Population Response (CPR) Model for Thermal Stress, Coastal and Estuarine Studies 61: Coral Reefs and Climate Change: Science and Management. Published by American Geophsical Union, Washington DC: p: 129-144 (2006).

3. van Woesik R, Lacharmoise F, Koksal S Annual cycles of solar insolation predict spawning times of Caribbean corals. Ecology Letters 9: 390-398 (2006).

4. Howel, B.P., Wood, S. and Kksal S., Passive Sonar Recognition and Analysis Using Hybrid Neural Networks, Oceans, 2004.

5. G. T. Bhaskar, S. Kksal and V. Lakshmikantham, Generalized Quasilinearization Method for Semilinear Hyperbolic Problems, Discrete and Continuous Dynamical Systems, pp. 1263-1275 (2003).

6. S. Kksal and V. Lakshmikantham, Unified Approach to Monotone Iterative Technique for Semilinear
Parabolic Problems, Dynamics of Continuous, Discrete and Impulsive Systems Series A,10(4):539-

V. Lakshimikantham and S. Kksal, Monotone Flows and Rapid Convergence for Nonlinear Partial Differential Equations, Francis and Taylor, 2003, London and New York.


Dr. Koksal is currently working on several projects that involve both undergraduate and graduate students:

  1. Neural Networks: Construction and analysis of Fuzzy backpropagation neural networks with vector valued activation functions, and application to data fusion.
  2. Biochemical Networks: Mathematical modeling of protein activities during oocyte maturation and fertilization. Nonlinear ordinary differential equations and neural networks approach are used for modeling.
  3. Coral Dynamics: Modeling coral population thresholds and vital rates using nonlinear difference and differential equations.

Research & Project Interests

Dr. Koksal’s research work and interests for the last 20 years can be summarized (in chronological order) as:

  1. Numerical solutions of magneto-hydro dynamic equations using the finite element method.
  2. Stability such as, uniform, asymptotic, exponential, eventual and practical, and boundedness properties of nonlinear ordinary differential equations by using Lyapunov's second method.
  3. The method of generalized quasilinearization and monotone iterative technique. Both of these methods provide an approximation for the solutions of nonlinear differential equations (both ordinary and partial) by means of lower and upper solutions. The main difference between these techniques is that generalized quasilinearization produces rapid convergence, namely, quadratic convergence that is of great interest in applications.
  4. Neural networks. Dr. Koksal has investigated various stability properties of continuous Hopfield neural networks and Grossberg competitive neural networks.
  5. Mathematical Biology.

In addition, she is interested in

  • Undergraduate mathematics education
  • Curriculum development
  • Research projects for undergraduates

Another one of her interests is to encourage young females to study math and pursue careers related to math. She is heavily involved with mentoring activities of Association of Women in Mathematics ( AWM).