Pipeline infrastructure and its safety are critical for the recovering of U.S. economy and our standard of living. Accurate pipe material strength estimation is critical for the integrity and risk assessment of aging pipeline infrastructure systems. Existing techniques focus on the single modality deterministic estimation of pipe strength/toughness and ignores inhomogeneousity and uncertainties. In view of this, a Bayesian network-based data analytics approach is proposed to accurately estimate the pipe material strength and toughness. This approach uses information fusion of multimodality surface measurement (e.g., surface chemistry, surface indentation and scratch testing results, and microstructure observations) to accurately predict the material strength. In addition, the proposed Bayesian network will be used to reduce the uncertainties in the material property estimation and enhance the confidence for operators and regulators for their decision making. Specifically, the proposed project will be in close collaboration with GTI. The developed relationships and database from GTI will be collected to develop the proposed Bayesian network for pipe materials (e.g., x42, x52, x60, x65, x70 and x80).