ORIGINAL PAPER Prediction of uniaxial compressive strength and elastic modulus of migmatites using various modeling techniques Bahman Saedi 1 & Seyed Davoud Mohammadi 1 & Hossein Shahbazi 1 Received: 30 January 2018 /Accepted: 10 September 2018 /Published online: 28 September 2018 # Saudi Society for Geosciences 2018 Abstract This study aims to develop several prediction models of uniaxial compressive strength (UCS) and elastic modulus (E) of different migmatite rocks from four areas of the Sanandaj-Sirjan zone in Iran. In addition to UCS and E, porosity, cylindrical punch Index (CPI), block punch index (BPI), Brazilian tensile strength (BTS), point load index (IS (50) ), and P wave velocity (V P ) were measured for migmatites. Various methods, like multiple regression (MR) analysis, artificial neural network (ANN), and adaptive neural fuzzy inference system (ANFIS), were used to predict UCS and E during the modeling process. In this study, a total of 120 inputs and outputs were used. According to the analyses performed in this study and the input parameters, five different models have been used to estimate UCS and E: (1) CPI, BPI, BTS, and IS (50) ; (2) CPI, BPI, BTS, and V P ; (3) CPI, BPI, IS (50) , and V P ; (4) CPI, BTS, IS (50) , and V P ; (5) BPI, BTS, IS (50) , and V P . Performance evaluation shows that ANN is a better prediction method compared to the others, and models 2, 4, and 5 are the best models for prediction. The developed models in this paper can have high prediction efficiency if they are used for similar types of rocks. Keywords Uniaxial compressive strength . Elastic modulus . Migmatite . Multiple regression . Artificial neural network . Adaptive neural fuzzy inference system Introduction Studying the engineering characteristics of rocks is one of the most important issues in engineering geology. Precise deter- mination of these characteristics makes it easy to decide on the construction of engineering structures and how to encounter various situations logically. In this regard, determining the engineering characteristics of rocks that have complicated be- havior is of utmost importance. Migmatites are heterogeneous rocks and are formed by subsequent processes. Migmatite is found in medium- and high-grade metamorphic areas and, on the macroscopic scale, is composed of two, or more parts with different petrography (Fig. 1). The parts created by partial melting are called neosome. Neosome usually contains pale- colored layers (leucosomes, composed of quartzofeldspathic, or feldspathic) and pitch-dark layers (melanosomes, enriched in ferromagnesian minerals). The part not affected by partial melting, which has a dark appearance and could be a parent rock, is called paleosome (mesosome) (Sawyer 2008). Uniaxial compressive strength (UCS) and elastic modulus of intact Rock (E) are of the most important mechanical prop- erties and are widely used in rock engineering, such as tunnel- ing, foundation design and construction, mining operations, slope stability studies, etc. (Karakus et al. 2005). Direct mea- surement of these characteristics in laboratory is intricate and time-consuming (Gokceoglu and Zorlu 2004; Baykasoglu et al. 2008; Jahed Armaghani et al. 2016a). Also, laboratory tests for determining UCS and E require expensive instru- ments (Eissa and Kazi 1988). To overcome this problem, var- ious indirect prediction models have been developed using simple test results or petrographic characteristics analysis (Bell 1978; Romana 1999; Sonmez et al. 2004; Tiryaki 2008). There are many empirical relationships to predict UCS and E (Kahraman 2001; Basu and Aydin 2006; Sarkar et al. 2012; Nazir et al. 2013; Liu et al. 2015). Usually, these correlations are not precise enough, while it is necessary to predict UCS and E precisely in rock engineering (Dehghan et al. 2010; Beiki et al. 2013). Therefore, several researchers have used multiple regression analysis (MR) to predict UCS and E * Seyed Davoud Mohammadi d.mohammadi@basu.ac.ir 1 Department of Geology, Faculty of Sciences, Bu-Ali Sina University, Mahdieh Ave, Hamedan 65178-38695, Iran Arabian Journal of Geosciences (2018) 11: 574 https://doi.org/10.1007/s12517-018-3912-9