This paper proposes using a decision contour derived from real options analysis, which is an evaluation tool for investment under uncertainty, to suggest an optimal stopping time of the compact genetic algorithm on the trap problem. The proposed criterion provides a stopping boundary, where termination is optimal on one side and continuation is on the other. A generic stopping function is formulated with an exercise region that scales well. The new stopping policy helps save on computational effort, and the evolutionary process reaches a higher solution quality when the reset method is incorporated. The proposed technique can be applied to analyze other problems.

This paper introduces the real options approach, which is an evaluation tool for investment under uncertainty, to analyze optimal stopping time in genetic algorithms. This paper focuses on the simple model of EDAs named the compact genetic algorithm. This algorithm employs the probability vector as a model that scales well with the problem size. We analyze optimal stopping time of trap problems and propose an optimal stopping criterion as a decision contour. The proposed criterion also provides a stopping boundary, where termination is optimal on one side and continuation is on the other. This region suggests when it is worth continuing the algorithm and helps save computational effort by stopping early. Moreover, when the reset method is applied, the algorithm can reach a higher solution quality. The proposed technique can also be applied to analyze other problems.

Software development projects are subject to external and internal risks that cause delays, budget overrun and poor quality. Portfolio management can be used to alleviate this problem, as it pools resources together and allows for resource sharing among projects. Consequently, projects are more likely to succeed. However, portfolio management using only deadlines and the number of employees to improve probability of success is still confined. This paper proposes integrating portfolio management with COCOMO II that offers more management flexibility. Managers can adjust other resources, such as tools, staff capability, communication support, etc. to improve the project's success. The proposed method can also be applied despite limited historical data and expert judgment. In addition, this paper introduces time constraints into portfolio management without assuming unrealistic linearity between effort and time.

Developing a complex system requires partitioning the target system into several subsystems. It is generally difficult to define each subsystem's scope and functional requirements as well as data dependencies among subsystems. Requirements Integration Model (RIM), which consists of a workflow model and a work procedure, can provide specific guidelines and techniques to increase quality of requirements obtained under a time constraint. We have applied the technique to the implementation of the educational system for the Department of Computer Engineering at Chulalongkorn University, Thailand. Experiences from this case study have shown that the model can assist developers to clearly specify functional requirements for each subsystem and to identify data dependencies among subsystems.

The real options technique has emerged as an evaluation tool for investment under uncertainty. It explicitly recognizes future decisions, and the exercise strategy is based on the optimal decisions in future periods. The real options approach has been applied to many economic and financial problems, but few are in computer science and engineering. The novelty of this work lies in applying real options to a computational problem. This paper proposes using the real options technique to find an optimal stopping decision for the compact genetic algorithm. The compact genetic algorithm, a kind of genetic algorithms, represents the population as a probability distribution over a set of solutions. This distribution automatically captures the underlying uncertainty of the problem, which can be simulated to obtain an evolutionary process of the algorithm. The experiments show preliminary results of employing the real options approach to determine the optimal stopping time for the compact genetic algorithm. The proposed technique can be applied to analyze other machine-learning algorithms, such as neural networks or other variations of genetic algorithms.

In this paper we present a prediction process of the Stock Exchange of Thailand index using adaptive evolution strategies. The prediction process does not require the knowledge of the functional form a priori. In each recursion step, genetic algorithm is used to evolve the structure of the prediction function, whereas the coefficient is evolved by evolution strategies. The proposed method has been shown to successfully predict the Stock Exchange of Thailand and returns an error less than 3%. This methodology is also a tool for knowledge discovery in a specific application. We have found that the SET index can be reasonably forecasted with only two factors: the Hang Seng index and minimum loan rate. The proposed method also achieves a lower prediction error when compared with multiple regression method