A Multi-Objective Approach to Integrated Multi-Mode Personnel Scheduling and Team Formation
Description
Description This dataset relates to a multi-skilled personnel staffing and activity scheduling problem that integrates workforce planning, activity scheduling, and team formation. The problem involves assigning workers to a set of activities that must be completed within a fixed planning horizon while considering skill requirements, activity durations, and worker availability. Each activity can be executed in one of several alternative modes, where each mode specifies a different duration and skill-based personnel requirement. The objective is to determine activity start times, select execution modes, and assign workers in order to minimize staffing costs while satisfying all activity requirements. In addition to cost efficiency, the dataset also considers team diversity as a performance measure. Each worker is assigned one of the 16 MBTI personality types, and the diversity of teams assigned to activities is evaluated using normalized interpersonal compatibility scores between personality types. This enables the study of optimization approaches that consider both operational efficiency and team composition. The instances are derived from benchmark data from the Project Scheduling Problem Library (PSPLIB). Activity durations and mode-dependent resource requirements are taken from PSPLIB instances, while additional information such as worker skills, personality types, and cost parameters is generated synthetically. Instance sets The dataset is organized into folders corresponding to different instance configurations. Each folder contains five instances. Three types of variations are considered. Activity size variation These datasets vary the number of activities: • j10 – 10 activities • j16 – 16 activities Both datasets include 3 modes per activity and 2 skill types. Mode variation These datasets vary the number of available execution modes per activity: • m1 – 1 mode • m2 – 2 modes • m4 – 4 modes • m5 – 5 modes Each dataset contains 16 activities and 2 skill types. Skill variation These datasets vary the number of skill types required by activities: • r1 – 1 skill • r3 – 3 skills • r4 – 4 skills • r5 – 5 skills Each dataset contains 16 activities and 3 modes per activity. Data format Each instance is provided as a plain text (.txt) file containing the input data required to solve the scheduling problem. The files include the following sections: • Sets: number of workers, skills, personality types, activities, and modes • Cost coefficients: worker deployment and duty costs • Activity characteristics: activity durations, modes, due dates, and skill-based resource requirements • Worker–skill matrix: binary matrix indicating the skills possessed by each worker • Worker personality types: binary matrix indicating the MBTI personality type assigned to each worker
Files
Institutions
- Ghent UniversityFlanders, Ghent