A 360-degree performance evaluation model using ant colony and particle swarm optimization algorithms: An application in a software company

Businesses must adopt continuous improvement and enhance their current situation to survive and maintain strong positions against competitors. The ability of businesses to progress towards their goals by improving their existing potential is only possible through the value and effort given to their employees. Therefore, businesses need to retain a qualified workforce that will provide a competitive advantage through various performance evaluation methods. The sustainability, growth, and ability to stand out in competitive market conditions of businesses are only possible by choosing the right and effective performance evaluation system. The 360-degree performance evaluation system, which businesses have started to use widely, allows employees to be evaluated multi-dimensionally by multiple evaluators, resulting in accurate and reliable outcomes. Providing feedback to employees based on the results of the 360-degree performance evaluation, analyzing their situations, and providing necessary support ensures an effective evaluation process. The innovative approach of the 360-degree performance evaluation system makes it more prominent than other methods. The benefits for businesses and employees and the wide usage areas of the evaluation results increase the usability of the system. However, alongside its advantages, some disadvantages indicate that the system can be improved. In this study, the evaluation of performance criteria determined by expert opinions for employees serving in a software company was examined within the 360-degree performance evaluation system model. The 360-degree performance evaluation system, as one of the modern methods, is distinguished from other classical methods. With this method, employees are evaluated in detail not only from a single perspective but also by subordinates, peers, managers, and customers. Based on the results of the 360-degree performance evaluation, action plans are prepared to provide information about the employees' situations and support in this direction. Implementing these action plans offers the opportunity for the business and the employee to continuously improve their performance in harmonious and efficient working conditions. The purpose of this performance evaluation model, prepared by applying the processes of the 360-degree performance evaluation system, is to design a sustainable system by accurately evaluating businesses and employees. For evaluation in the software company, team leaders, senior consultants, and consultants serving in the company were selected. The organization consists of team leaders, senior consultants, and consultants. Senior consultants manage a group of consultants reporting to them. Since consultants are new to the company and do not have as much experience as senior consultants, they do not have a group reporting to them. To evaluate these employee groups, a competency group defined within the 360-degree performance evaluation system was prepared based on expert opinions, including core functional and managerial competencies. Core competencies are criteria that cover the general statusof employees in the business. Functional competencies are criteria observed based on the employees' knowledge and experience related to the profession. Managerial functions are criteria where the management activities of employees in managerial positions are evaluated. Competencies defined in the 360-degree performance evaluation system are seen to be usable in evaluations as they are tied to the business's mission and vision, suitable for objective evaluation, and comprehensible. In practice, to select the criteria in order of importance and to find the weights of these criteria in the performance evaluation system, metaheuristic methods such as the Ant Colony Algorithm and Particle Swarm Optimization Algorithm were used. These heuristic methods, based on swarm intelligence, solve many optimization problems by taking examples from the behaviors of living beings in nature. While the ant colony algorithm is generally used for scheduling and pathfinding problem optimization, this study showed that it can also be used for selecting competencies in order of importance. Particle swarm optimization is commonly used in studies for the optimization of routing and traveling salesperson problems. In this study, with the data integrated from the ant colony algorithm, finding the criteria weights brought a new perspective to the criteria weight finding problems. Correct selection of the input parameters for the ant colony to reach the best result is important. Therefore, the most suitable values for the algorithm parameters were selected, and the results from the system were analyzed with error margins. In particle swarm optimization, the objective function optimizing the criteria weights in the evaluation system was defined, and the most suitable parameter values were given to the algorithm. The system's results were analyzed with error margins by running the Particle Swarm Optimization according to the parameter values. According to the error margin measurements, the iteration with the least error margin was selected and taken as data for the 360-degree performance evaluation system. Having too many competencies to evaluate in the 360-degree performance system would negatively affect the system's performance. To ensure the system works smoothly and the evaluation does not take too much time, decision-makers were asked to rate the important competencies to select from the competency groups. The most important competencies were gathered and run in the ant colony algorithm according to the decision-makers' ratings. The error margins of the results from the ant colony algorithm were measured. To minimize the error margin and make the results close to reality, the Ant Colony Algorithm was run five times, and the algorithm results were tested. The iteration with the least error margin in the Ant Colony Algorithm provided the criteria to be used in the application. After selecting the criteria to be used in the application, the Particle Swarm Optimization was run ten times to find the weights of the criteria. The error margins of the criteria weights were measured by mean absolute error (MAE), mean squared error (MSE), and mean absolute percentage error (MAPE). The iteration with the lowest error margin was determined as the criteria weights in the algorithm results. The method applied in the study, using the two algorithms in an integrated manner, brought a new approach to the 360-degree performance evaluation system. With the method used in the application, the aim was to find the best algorithm providing the optimization of criteria weights and the selection of criteria in order of importance. Team leaders, senior consultants, and consultants are evaluated by subordinates, peers, managers, and customers. Since there is an organizational structure reporting to the team leader, the team leaders are evaluated by subordinates, peers, managers, andcustomers. The evaluation rates of each evaluator can change as criteria weights since the status of each evaluator differs. Senior consultants are evaluated by subordinates, peers, managers, and external customers, as there is a group of consultants reporting to them. Since there is no organization reporting to the consultants, they are only evaluated by peers, managers, and customers. The criteria weights for the evaluator categories of team leaders, senior consultants, and consultants were found using Particle Swarm Optimization. The weights of the three basic competency categories defined in the system are not the same for team leaders, senior consultants, and consultants. The competencies evaluated for team leaders are core, functional, and managerial competencies. The competencies evaluated for senior consultants are core, functional, and managerial competencies. Consultants are evaluated only in core and functional competency groups. Particle Swarm Optimization was used to find the criteria weights of competencies according to titles. The 360-degree performance evaluation system was integrated into a software program, and the competencies, evaluated groups, and evaluators were defined in the system. This software program was designed as a sample application. In this model, team leaders evaluated by subordinates, peers, managers, and customers receive a score based on competencies. Additionally, the feedback reports of the evaluation results in the system allow the team leader to see their strengths and weaknesses. In the 360-degree performance evaluation system, the evaluators remain confidential to obtain impartial and reliable results. Objective evaluation free from bias facilitates the smooth operation of the system. Managers decide on the necessary training plans and programs for the business based on the evaluation results and feedback reports. Adopting the 360-degree performance evaluation results and supporting the applied training increases the commitment to the system. Based on the most suitable evaluation data in this model, an effective performance evaluation system for the company was proposed.