Search

Task allocation is a critical phase of project management. Tree-type structures are frequently used constraints to obtain a pertinent task allocation. They can illustrate where one task may require numerous agents and when an agent can be assigned to different tasks (roles). The process of task allocation is made more complex when administrators need to satisfy sequential and fixed branch relationships between/among tasks (roles). This paper formalizes the tree-structured task allocation problem (TSTAP) with group multirole assignment (GMRA) and proves necessary conditions, the necessary and sufficient condition, as well as sufficient conditions, of TSTAP. The formalization makes it easy to find a solution with the IBM ILOG CPLEX optimization package (CPLEX). The necessary conditions improve the CPLEX solution by eliminating infeasible cases. The necessary and sufficient condition describes the solution space of TSTAP completely. Another exciting result is that the sufficient conditions can not only improve the CPLEX solution by describing a practical approximate solution space but also help decision-makers and human resource officers organize a team in order to successfully assign tasks. The proposed approach is verified by simulation experiments with respect to a real-world problem. The experimental results present the practicability of the proposed solutions in this paper. This paper was motivated by general cooperative projects whose tasks have tree-structured relationships. This can make the problem of successful multitask assignment extremely challenging. The traditional method of assignment such as the KM algorithm can no longer solve this problem. To solve the assignment problem with tree-structured relationships, an efficient many-to-many assignment with constraints is required. The proposed approach provides theoretical and technical foundations for efficient assignment of TSTA, which can not only provide a viable and effective assignment scheme for TSTA problems but also help human resource officers to formulate reasonable plans according to the relationships between/among tasks. [ABSTRACT FROM AUTHOR]

Role assignment is a critical task in role-based collaboration. It has three steps, i.e., agent evaluation, group role assignment, and role transfer, where group role assignment is a time-consuming process. This paper clarifies the group role assignment problem (GRAP), describes a general assignment problem (GAP), converts a GRAP to a GAP, proposes an efficient algorithm based on the Kuhn–Munkres (K-M) algorithm, conducts numerical experiments, and analyzes the solutions' performances. The results show that the proposed algorithm significantly improves the algorithm based on exhaustive search. The major contributions of this paper include formally defining the GRAPs, giving a general efficient solution for them, and expanding the application scope of the K-M algorithm. This paper offers an efficient enough solution based on the K-M algorithm that outperforms significantly the exhaustive search approach. [ABSTRACT FROM PUBLISHER]

Many-to-Many (M-M) role transfers are generalized problems that are encountered in collaboration. Exhaustive-search-based algorithms are too computationally intensive. This paper introduces the Kuhn–Munkres (or Hungarian) algorithm for the general assignment problems (GAPs) and proposes a new efficient algorithm to solve the M-M role transfer problems by converting them to the GAPs. The experiments and results validate the proposed algorithms. [ABSTRACT FROM PUBLISHER]

Abstract: Web services collaborations are challenging as research and practical tasks. They are highly automatic, dynamic, heterogeneous, and lack protection against corruption of processes. High risks are imposed on the services (including the consumers and providers) in collaboration. Hence, avoiding illegal collection of private data disclosed in services collaboration is becoming increasingly important. This paper studies how to realize the minimal privacy disclosure while achieving the functional objectives through role mechanisms. Initially, it proposes a revised role-based framework for privacy-aware services collaborations, which considers the impact on the reputation degree of the historic experiences of services in playing roles. Next, it models the privacy behaviors of services by extending the interface automata to support privacy semantics. Furthermore, it quantitatively analyzes the minimal privacy authorization; and presents the minimal privacy delegation algorithm, which allows us to automatically derive optimal privacy policies for an cross-organizational services collaborations system. Finally, it verifies the correctness and efficiency of this role-based approach through a case study. [ABSTRACT FROM AUTHOR]

Many-to-many (M-M) role transfers are generalized problems that occur normally in collaboration, management, and the operation of organizations. This paper extends our previous work on role transfer. It is the first attempt to formalize the M-M role-transfer problems with real-world management examples, analyze the different cases of M-M problems, design algorithms for all the cases, implement all the algorithms to provide solutions, and discuss the future work. The significant contribution of this paper is to demonstrate that role-transfer problems have definite solutions. The implementations and their results validate the proposed algorithms. [ABSTRACT FROM AUTHOR]