Government office service provision is being supported by networking infrastructure to connect their branch office with the central datacenter from which the service of application and database is accessed and shared. MoR is among the main government institution which is legally established to levy and collect the inland and custom taxes from the concerned bodies. MoR network dropouts is noticed and become the most discussed issues not only with management and Ethio telecom officials but also with taxpayers as clearly shown on annual reports, stakeholders meeting, and ten years strategic plan. rnCurrently Ethiopian tax management is being supported by systems developed to support the taxation process based on the defined laws and regulations. The Branch offices are connected to the main datacenter to share the central resources. Though MoR is continually upgrading bandwidth and replacing its devices that are assumed to cause network dropouts, the problem of network dropouts is still unresolved and become the main cause of taxpayers complains publicly. This thesis is aimed to analyze MoR network infrastructure to find the major causes for network dropouts, which leads to taxpayer dissatisfaction with the service provided by the organization. To accomplish the main objective of the thesis MoR network infrastructure is analyzed based on the primary and secondary data collected from network administrators and the monitoring tools installed in the datacenter. The collected data from the SolarWinds and PRTG is analyzed based on the metrics availability, utilization, error and discards, delay, and protocol overhead. rnThe simulation of the existing network infrastructure is done to further understand the network problems causing network congestion and system unavailability by using OPNET simulation tools, the simulation also help us to demonstrate and test recommended solutions. The analysis result leads us to conclude that there is underperformance caused by different factors like structural imbalance, overloads, lack of system tuning and protocol overheads, on all stages of MoR network and this leads to experiencing some problems like unhealthy link and resource utilization, packets discards and error, high device response time, ISP service availability problems. The major causes are lack of system tuning, overloads, not establishing enterprise-wide performance metrics or improvement targets and policies, not having complete visibility into the network to efficiently manage and operate the enterprise.Then optimization framework consisting 6 modules is proposed which can withstand the problem noticed in the previous stages. The building blocks of the rnrnframeworks are tested on the OPNET network modeling tools and the result shows improvement on the configured metrics. rnFinally, to overcome the network problems this thesis highly recommends toward MoR management to consider the findings of this paper and implementing the framework. MoR also needs to establish an IT policy and a network strategy to include a process that continuously monitor IT infrastructures, and develop enterprise-wide performance metrics. Additionally the organization need to establish improvement targets in coordination with the management to implement a solution, such as site wide emergency generators that is fully dedicated only for the data centre to avoid relaying only on the building power generator, this will prevent network connectivity and availability issues.