Introduction: Rice (Oryza Sativa), a member of Gramineae family, is one of the most important cereal grains which can be annual or perennial. The grain has many varieties which are divided into Asian and African classes. Rice grain and its related products are food staple of over 40% of global population (Sereewatthanawut et al. 2008). Rough rice (paddy) is surrounded by a loose outer hull or husk. The white starchy endosperm kernel of rice is enclosed by a firmly adhering bran coat and germ. In general, rice is milled before utilization, producing hull, bran, germ, and white rice. Rice bran, process discard, is extensively used as animal feed, fuel or fertilizer throughout the world (Saunders 1985). Rice bran is rich in protein (12-16%), fat (16-22%), raw fiber (8-12%) as well as vitamins and minerals (e. g., thiamin, niacin, aluminum, chlorine, iron, magnesium, phosphorus, potassium, silicon, sodium, and zinc). This byproduct is good source of vitamin E such as tocopherols (alpha, beta, sigma, and gamma). Numerous studies have shown the beneficial effects of vitamin E in the prevention, control, and treatment of various disease. Moreover, over the years many studies have confirmed the health benefiting effects of rice bran (Ryan et al. 2011). Recently, immunoregulatory effects of rice bran, against enteric pathogens, have been reported (Kumar et al. 2012; Nealon et al. 2019). Rice bran, as a fermentable substrate for the growth of the probiotic microorganisms, has been successfully used for the production of functional foods (Poletto et al. 2019). To the best of our knowledge, there are no reports on the use of rice bran in the formulation of Fouman cookie. Thus, the present study was designed: (a) to examine the effects of rice bran on the chemical, physical and organoleptic properties of Fouman cookie; (b) to obtain the relationship between Fouman cookie parameters using multivariate mathematical-statistical methods such as partial least squares regression analysis (PLSR) and principal components analyses (PCA) as ways to offer simple ways to measure texture, sensory and shelf-life aspects. Material and methods: The stabilized rice bran was kindly provided by the Giltaz Company (Langarud, Guilan, Iran). The obtained bran powder was stored at-18℃ till further analysis. In this research, we investigated the effect of different levels of rice bran (0, 10% and 20%) in production of Fouman cookie. Moisture, protein, ash, fat, total dietary fiber, phytic acid, acidity, peroxide, and vitamins (thiamin, niacin, B6, E and pantothenic acid) were analyzed by method of American Association for Clinical Chemistry (AACC (2000). The prepared samples were coded by two-digit codes and presented in random order to 10 trained evaluators. They were asked to score the cookies in the range of 1 to 5 in terms of odor, taste, color, texture, and overall quality (5-point hedonic scale). All the assessments were conducted on the same day at room temperature. A texture analyzer (Brookfield, Model CT310K, US) was used to measure the force required for compression of a round-bottom (2. 5 cm diameter × 1. 8 cm height) probe on the samples. The test velocity was set at 0. 5 mm/s with a compression depth of 5 mm and the trigger load of 100 g. In a typical process, the samples were placed vertically inside the device and the probe landed on it with the mentioned velocity and compression depth (El-Arini and Clas 2002). The texture of samples was evaluated after 1, 72 and 168 hours of storage for shelf-life monitoring. Crumb texture was determined by a texture analyzer (Brookfield, Model CT310K, US) provided with the software “ Texture Expert” . The aluminum cylindrical probe (25 mm diameter) was used in a “ Texture Profile Analysis” double compression test (TPA) to penetrate to 50% depth, at 2 mm/s speed test, with a delay of 30 s between first and second compression. Hardness (N), springiness and resilience were calculated from the TPA graphic (Gó mez et al. 2008). A central region image of each sample was captured using a flatbed HP Scanjet G4010 Photo Scanner (Hewlett-Packard, Palo-Alto, CA, USA) supporting Desk Scan II software (Hewlett Packard, USA). A single 8 mm×8 mm square field of view was analyzed for each image. Brightness was adapted to150 units and contrast to 170 units. Images were scanned in 256 grey levels at 150 dots per inch (dpi) each containing 355 columns by 355 rows of picture elements (pixels) (Crowley et al. 2002). ImageJ 1. 4g (National Institute of Health, USA) was used for analyzing of the JPEG image file. The CIE L*a*b* (or CIELAB) color model was operated for the examination of color (Figure 1). The three parameters of such model represent the color lightness (L*), redness-greenness (a*) and blueness-yellowness (b*) (Pourfarzad and Habibi-Najafi 2012; Quevedo et al. 2009). To evaluate crumb structure, color images were converted to 8-bits 256 gray level images. The thresholding method (conversion to a binary image) of the 256 gray level digital images was used for image segmentation. The selected crumb grain features were the average size and porosity (FarreraRebollo et al. 2011, Angioloni and Collar 2011). To assess the significant differences among samples, a completely randomized design was performed (Minitab 15, Minitab Inc., State College, PA, USA). To investigate the significance of differences between the mean values, Duncan’ s multiple range tests were employed after the one way analysis of variance at the confidence level of 95%. Each test was conducted in three replicates. Principal component analysis and PLSR were performed on physicochemical, texture and sensory data sets. Results and discussion: The results showed that there was no significant difference between the color score, texture and the overall quality of control and samples containing 10% bran, but their scores decreased significantly with increasing bran amount up to 20%. The moisture and ash contents of the samples increased significantly with increasing rice bran amount. Also, by adding rice bran, the parameters of crumb L, crust L, crust b, porosity, and average size of Fouman cookie significantly decreased. Moreover, addition of rice bran increased the hardness of samples on the first day, but there was no significant difference between the hardness of the control and samples containing10% rice bran during storage. Conclusion: The best physico-chemical, sensory properties and shelf-life of Fouman cookie were obtained at 10% fortification level. Obtained Partial least squares regression models had very high R 2 indicating the high efficiency of these equations in predicting the sensory properties of Fouman cookies using color analysis indices.