This paper introduces a network Data Envelopment Analysis (DEA) model based on the slacks-based measure (SBM) approach to evaluate divisional and overall efficiencies in organizations with internal linking activities. Traditional DEA models neglect intermediate products or linking activities, which are crucial for accurately measuring efficiency in multi-divisional organizations. The proposed network DEA model accounts for these intermediate activities and allows for the evaluation of divisional efficiencies as well as overall efficiency. The model uses the SBM approach, which is a non-radial method suitable for measuring efficiencies when inputs and outputs may change non-proportionally. It decomposes overall efficiency into divisional efficiencies and incorporates the importance of divisions through a weighted SBM model. The model is applied to various real-world scenarios, including electric power companies, hospitals, broadcasting companies, and financial holding companies, where internal linking activities are present. The paper discusses several properties of the network DEA model, including the relationship between overall and divisional efficiencies, the existence of divisionally efficient DMUs under different assumptions (variable returns-to-scale and constant returns-to-scale), and the efficiency of projected DMUs. It also presents illustrative examples, including a case where no divisionally efficient DMUs exist under constant returns-to-scale and free link assumptions. The model is extended to incorporate link flows into efficiency measurements and explores the role of the intensity vector in the model. The paper concludes that the proposed network DEA model provides a more accurate and comprehensive evaluation of efficiency in organizations with complex internal structures compared to traditional DEA models. It also highlights the importance of considering intermediate activities and the need for further research in this area.This paper introduces a network Data Envelopment Analysis (DEA) model based on the slacks-based measure (SBM) approach to evaluate divisional and overall efficiencies in organizations with internal linking activities. Traditional DEA models neglect intermediate products or linking activities, which are crucial for accurately measuring efficiency in multi-divisional organizations. The proposed network DEA model accounts for these intermediate activities and allows for the evaluation of divisional efficiencies as well as overall efficiency. The model uses the SBM approach, which is a non-radial method suitable for measuring efficiencies when inputs and outputs may change non-proportionally. It decomposes overall efficiency into divisional efficiencies and incorporates the importance of divisions through a weighted SBM model. The model is applied to various real-world scenarios, including electric power companies, hospitals, broadcasting companies, and financial holding companies, where internal linking activities are present. The paper discusses several properties of the network DEA model, including the relationship between overall and divisional efficiencies, the existence of divisionally efficient DMUs under different assumptions (variable returns-to-scale and constant returns-to-scale), and the efficiency of projected DMUs. It also presents illustrative examples, including a case where no divisionally efficient DMUs exist under constant returns-to-scale and free link assumptions. The model is extended to incorporate link flows into efficiency measurements and explores the role of the intensity vector in the model. The paper concludes that the proposed network DEA model provides a more accurate and comprehensive evaluation of efficiency in organizations with complex internal structures compared to traditional DEA models. It also highlights the importance of considering intermediate activities and the need for further research in this area.