Development of a Heuristic Based Mixed Integer Linear Programming Model for Resources Allocation During Cyberfraud Mitigation

In this digital era, cyberfraud remains one of the major challenges faced by individuals and financial institutions. Lack of suitable decision support systems for effective allocation of resources affects the success rate in combating the crime. Thus, the purpose of this study is to develop a mixed integer linear programming model that can assist financial institutions to make proactive decisions about resource allocation in their response to cyberthreats using the South African banking industry as a case study. To achieve this, literature review was carried out and the components of an organisation’s cyber-incidence response plans were established. These are the following: cyber-incidence identification, detection, protection, response, and recovery. Mathematical formulations for the optimisation of the cost of resource allocation and connections among the banking head offices, branches and automated teller machine (ATM) outlets were established. The formulation was carried out using mixed-integer linear programming (MILP). The heuristic algorithm found the optimal objective function value of $$2.4382\times\,{10}^{6}$$ 2.4382 × 10 6 at the 13th iteration at node 0 after exploring 32 nodes for the first objective while the algorithm found the optimal objective function value of 3872.00 at the 120th iteration at node 50 after exploring 55 nodes for the second objective function. The results obtained from the MILP model indicated that the heuristic algorithm can be used for the ordering, sequencing, and assignment of tasks to the antifraud teams at minimum costs. It can also be used to establish the best possible connection among these facilities to ensure a quick incidence response during cyberattack. Hence, the implementation of the MILP model for cost minimisation and effective allocation and utilisation of human resources to combat cyberfraud is recommended.