Document Type : Research Paper

Authors

• Mehri Chehrehpak ¹
• Abbas Tolouei Ashlaghi ²
• Kamran Mohammadkhani ³

¹ Ph.D Candidate, Management and Economy Faculty, Science and Research Branch, Islamic Azad University, Tehran, Iran

² Professor, Management and Economy Faculty, Science and Research Branch, Islamic Azad University, Tehran, Iran Corresponding Author: toloei@srbiau.ac.ir

³ Professor, Management and Economy Faculty, Science and Research Branch, Islamic Azad University, Tehran, Iran

Abstract

Effective knowledge-based processes are essential for companies operating in the information technology industry. These
Effective knowledge-based processes are essential for companies operating in the information technology industry. These processes rely on the expertise of skilled workers and play a crucial role in the value chain of such organizations. Decision-making is a critical element of knowledge-based processes, highlighting the need to identify decision rules and models accurately. In this paper, we examine the process of identifying and deciding on proposed ideas in the software industry, analyzing decision logs from a leading software company. The Rough sets theory and fast Reduction algorithm are employed to provide a step-by-step approach to data analysis and decision mining. The algorithm identifies vital features used in decision-making and presents the decision model as if-then rules, utilizing existing equivalence rules between data. The results demonstrate that this model can significantly reduce the direct involvement of decision-makers and the duration of the decision-making process. In today's competitive landscape, effective knowledge-intensive processes are fundamental for companies in the information technology (IT) industry. These processes are highly dependent on the expertise of skilled professionals and are integral to value creation across various organizational fronts. Decision-making—considered a cornerstone of knowledge-intensive processes—underscores the necessity of accurately identifying decision rules and models. This paper focuses on the methods of identifying and evaluating proposed ideas within the software industry, specifically analyzing decision logs from a leading software company. By employing the Rough Set Theory along with the Fast Reduction Algorithm, we provide a detailed methodological framework for data analysis and decision mining. This structured algorithm identifies critical features relevant to decision-making and presents the resulting decision model in the form of if-then rules, which are derived from pre-existing equivalence relations among data. Our results illustrate that the implemented model can significantly lessen the direct involvement of decision-makers as well as the time taken in the decision-making process, highlighting a potential path for enhancing operational efficiency in IT firms.

Introduction

The field of information technology is constantly evolving, marked by rapid developments and intense competition. To navigate this landscape successfully, organizations must rely on effective knowledge-based processes that are essential for sustaining competitive advantages. These processes hinge on the expertise of skilled workers who play a pivotal role in various stages of product development and innovation.
This paper aims to illuminate the decision-making facets of knowledge-intensive processes in the context of new idea generation within software companies. By scrutinizing decision logs from a prominent software firm, we aspire to discern decision rules and models that could significantly optimize decision-making efficiencies, ultimately positively impacting innovation outcomes.

Research Questions

This research is driven by several key inquiries aimed at uncovering various dimensions of decision-making in IT innovation processes:

What methods can be employed to identify decision points in the innovation processes of IT companies?This question targets the analytical techniques used to pinpoint where crucial decisions occur during the innovation lifecycle.
How can critical decision-making features be identified within these organizations, and what are the characteristics of these features?Identifying these features assists in understanding what influences decisions, including both internal and external factors.
In what ways can structured procedures be developed to expedite and improve the decision-making processes in IT innovation?This question seeks to establish procedural guidelines that can streamline decision-making, allowing companies to react swiftly to new information and emerging market trends.


Literature Review

The importance of Business Process Management (BPM) and decision mining in enhancing organizational efficiency is well documented in the literature. Earlier studies have primarily focused on implementing process mining techniques across various sectors, including healthcare and manufacturing, to improve overall decision-making efficiency. However, there exists a relative scarcity of research that specifically addresses decision mining in the context of IT innovation processes.
This study builds on existing frameworks, particularly leveraging the Rough Set Theory and the Fast Reduction Algorithm. These methodologies facilitate a thorough analysis of decision-making features, enabling the development of a tailored decision model for the software industry. By filling this notable gap, our research generates insights that can be applied to enhance decision-making within knowledge-intensive sectors.

Methodology

This research employs a comprehensive case study methodology, focusing on a well-established Iranian IT firm with over 25 years of industry experience. Our approach is structured into several key phases:

Identifying Decision Points: We apply a four-stage model, as outlined by Bazhenova and Weske (2016), to systematically pinpoint decision-making instances throughout the innovation process.
Analyzing Decision Logs: In this phase, we extract and scrutinize decision logs to identify critical features that influence decision-making. This analysis involves various statistical and data mining methods to validate findings.
Utilizing Rough Set Theory and Fast Reduction Algorithm: Following feature extraction, we employ Rough Set Theory alongside the Fast Reduction Algorithm to develop a robust decision model. This model is articulated through if-then rules that encapsulate significant decision-making aspects.
Evaluating Model Effectiveness: To ascertain the model's effectiveness, we conduct an extensive analysis of the product development process within the company, assessing how well the model predicts decision outcomes.


Results

The results of implementing the proposed decision model revealed several significant features critical to decision-making processes:

Idea Relevance: The relationship of the proposed idea to existing business operations emerged as a crucial factor.
Idea Source: Determining whether the idea originated from internal staff or external consultants significantly influenced the decision-making progression.
Anticipated Customer Acceptance: Factors related to customer acceptance and assessments of the competitive landscape were also primary considerations in the decision-making process.

The model showcased a remarkable 91.5% accuracy rate in predicting decision outcomes based on the identified features, illustrating its effectiveness. More importantly, the implementation resulted in a pronounced reduction in the direct involvement of decision-makers and a considerable decrease in the duration required for decision-making processes.

Conclusion

The research findings underscore the potential of applying Rough Set Theory along with decision mining techniques to significantly enhance the efficiency of decision-making in IT innovation processes. By systematically identifying and modeling essential decision features, organizations can streamline operations, minimize redundant tasks, and improve the overall effectiveness of their innovation strategies.
This study contributes to the growing body of knowledge on decision mining in the software industry, offering a structured approach that can be adapted to various knowledge-intensive environments. Looking ahead, further research is needed to explore the adaptability of this model in larger organizations and diverse contexts, further expanding its applicability within the broader IT landscape.
The implications of this research extend beyond the immediate findings, suggesting that strategic implementation of structured decision-making models can enhance operational efficiency across various sectors. Future studies could investigate the scalability of these models in larger organizations and their applicability in other innovation-driven industries.
Keywords: Process Mining, Decision Mining, Rough Set Theory, Knowledge-Intensive Process, Information Technology.
 
 
 
 
 

Keywords

• Process Mining
• Decision Mining
• Rough Set Theory
• Knowledge-intensive Process
• Information Technology

Main Subjects

• Data, information and knowledge management in the field of smart business