Hande KÜÇÜKAYDIN, Berkan HÖKE, Zeynep TURGAY, Cem ÜNSALAN

Determining and evaluating new store locations using remote sensing and machine learning

Decision making for store locations is crucial for retail companies as the profit depends on the location. The key point for correct store location is profit approximation, which is highly dependent on population of the corresponding region, and hence, the volume of the residential area. Thus, estimating building volumes provides insight about the revenue if a new store is about to be opened there. Remote sensing through stereo/tri-stereo satellite images provides wide area coverage as well as adequate resolution for three dimensional reconstruction for volume estimation. We reconstruct 3D map of corresponding region with the help of semiglobal matching and mask R-CNN algorithms for this purpose. Using the existing store data, we construct models for estimating the revenue based on surrounding building volumes. In order to choose the right location, the suitable utility model, which calculates store revenues, should be rigorously determined. Moreover, model parameters should be assessed as correctly as possible. Instead of using randomly generated parameters, we employ remote sensing, computer vision, and machine learning techniques, which provide a novel way for evaluating new store locations.

PDF

___

[1] Daskin M. Network and discrete location: models, algorithms and applications. Journal of the Operational Research Society 1997; 48 (7): 763–764.
[2] Küçükaydın H. Optimally Locating Facilities with Variable Characteristics. PhD, Boğaziçi University, Istanbul, Turkey, 2011.
[3] Farahani RZ, Hekmatfar M. Facility Location: Concepts, Models, Algorithms and Case Studies. Heidelberg,Germany: Springer, 2nd edition, 2009.
[4] Mendes AB, Themido IH. Multi-outlet retail site location assessment. International Transactions in Operational Research 2004; 11 (1): 1–18.
[5] Roig-Tierno N, Baviera-Puig A, Buitrago-Vera J, Mas-Verdu F. The retail site location decision process using gis and the analytical hierarchy process. Applied Geography 2013; 40: 191–198.
[6] Drezner T. Locating a single new facility among existing, unequally attractive facilities. Journal of Regional Science 1994; 34 (2): 237–252.
[7] Plastria F, Carrizosa E. Optimal location and design of a competitive facility. Mathematical Programming 2004; 2: 247–265.
[8] Pelegrín B, Fernández P, García Pérez MD. Profit maximization and reduction of the cannibalization effect in chain expansion. Annals of Operations Research 2016; 246: 57–75.
[9] Reilly WJ. The Law of Retail Gravitation. New York, USA: Knickerbocker Press, 1931.
[10] Huff DL. Defining and estimating a trading area. Journal of Marketing 1964; 28 (3): 34–38.
[11] Huff DL. A programmed solution for approximating an optimum retail location. Land Economics 1966; 42 (3): 293–303.
[12] Nakanishi M, Cooper LG. Parameter estimation for multiplicative competitive interaction model: least squares approach. Journal of Marketing Research 1974; 11 (3): 303–311.
[13] Achabal DD, Gorr WL, Mahajan V. Multiloc: A multiple store location decision model. Journal of Retailing 1982; 58 (2): 5–25.
[14] Aboolian R, Berman O, Krass D. Competitive facility location and design problem. European Journal of Operations Research 2007; 182 (1): 40–62.
[15] Hall RW. Handbook of Transportation Science. Kluwer Academic Publishers, 1999.
[16] Abouee-Mehrizi H, Babri S, Berman O, Shavandi H. Optimizing capacity, pricing and location decisions on a congested network with bulking. Mathematical Methods of Operations Research 2011; 74: 235–255.
[17] Küçükaydın H, Aras N, Altınel IK. A discrete competitive facility location model with variable attractiveness. Journal of the Operational Research Society 2011; 62 (9): 1726–1741.
[18] Aboolian R, Berman O, Krass D. Efficient solution approaches for a discrete multi-facility competitive interaction model. Annals of Operations Research 2009; 167: 297–306.
[19] Drezner T, Drezner Z. Lost demand in competitive environment. Journal of the Operational Research Society 2008; 59 (3): 362–371.
[20] Benati S, Hansen P. The maximum capture problem with random utilities: Problem formulation and algorithms. European Journal of Operational Research 2002; 143 (3): 518–530.
[21] Bello L, Blanquero R, Carrizosa E. On minimax-regret huff location models. Computers and Operations Research 2011; 38 (1): 90–97.
[22] Drezner T, Drezner Z. Validating the gravity-based competitive location model using inferred attractiveness. Annals of Operations Research 2002; 111: 227–237.
[23] Drezner T, Drezner Z, Zerom D. Competitive facility location with random attractiveness. Operations Research Letters 2018; 46 (3): 312–317.
[24] Ashtiani MG. Competitive location: a state-of-art review. International Journal of Industrial Engineering Computations 2016; 7 (1): 1–18.
[25] Hartley R, Zisserman A. Multiple View Geometry in Computer Vision. Cambridge, UK: Cambridge University Press, 2nd edition, 2004.
[26] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems 2012; 25: 1097–1105.
[27] Girshick R, Donahue J, Darrell T, Malik J. Region-based convolutional networks for accurate object detection and segmentation. IEEE transactions on pattern analysis and machine intelligence 2015; 38 (1): 142–158.
[28] Girshick R. Fast r-cnn. In IEEE International Conference on Computer Vision (ICCV). Santiago, Chile, 2015; pp. 1440–1448.
[29] Ren S, He K, Girshick R, Sun J. Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 2017; 39 (6): 1137–1149.
[30] He K, Gkioxari G, Dollar P, Girshick R. Mask r-cnn. In Proceedings of the IEEE International Conference on Computer Vision (ICCV). Venice, Italy, 2017; pp. 2961–2969.
[31] Abdulla W. Mask r-cnn for object detection and instance segmentation on keras and tensorflow. https://github.com/matterport/Mask_RCNN, 2017.
[32] Birchfield S, Tomasi C. Depth discontinuities by pixel-to-pixel stereo. International Journal of Computer Vision 1999; 35 (3): 269–293.
[33] Hirschmuller H. Accurate and efficient stereo processing by semi-global matching and mutual information. In IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05). San Diego, CA, USA, 2005; pp. 807–814.
[34] Rupnik E, Daakir M, Deseilligny MP. Micmac–a free, open-source solution for photogrammetry. Open Geospatial Data, Software and Standards 2017; 2 (1): 14.
[35] Lowe DG. Distinctive image features from scale-invariant keypoints. International journal of computer vision 2004; 60 (2): 91–110.
[36] Indyk P, Motwani R. Approximate nearest neighbors: Towards removing the curse of dimensionality. In Proceedings of the Thirtieth Annual ACM Symposium on Theory of Computing, STOC ’98. Association for Computing Machinery, New York, NY, USA, 1998; p. 604–613.
[37] Press WH, Teukolsky SA, Vetterling WT, Flannery BP. Numerical Recipes: The Art of Scientific Computing. New York, USA: Cambridge University Press, 1986.
[38] Talbi EG. Metaheuristics: From Design to Implementation. Hoboken, NJ, USA: John Wiley & Sons, 2009.
[39] Drezner Z. Facility location: a survey of applications and methods, volume 1. New York, USA: Springer-Verlag, 1995.
[40] Drezner T. Derived attractiveness of shopping malls. IMA Journal of Management Mathematics 2006; 17 (4): 349–358.