Journal of Statistical Planning and Inference 136 (2006) 2805 – 2819 www.elsevier.com/locate/jspi On multi-level supersaturated designs S. Georgiou, C. Koukouvinos ∗ , P. Mantas Department of Mathematics, National Technical University of Athens, Zografou, 15773 Athens, Greece Received 19 November 2003; accepted 8 November 2004 Available online 24 December 2004 Abstract In this paper we present a method of construction E(f NOD )-optimal multi-level supersaturated design with n rows, m columns and the equal occurrence property, from a resolvable balanced in- complete block design. A connection between orthogonal arrays and resolvable balanced incom- plete block designs is discussed and some E(f NOD )-optimal multi-level supersaturated designs are provided. © 2004 Elsevier B.V.All rights reserved. MSC: primary 62K10; 62K15; secondary 05B20 Keywords: Supersaturated designs; Factorial designs; Resolvable balanced incomplete block designs; Orthogonal arrays; Dependency; Efficiency 1. Introduction Supersaturated design is used in the initial stage of an industrial or scientific experiment for screening the active factors, and is useful when there are a large num- ber of factors under investigation while only a very limited number of experimental runs is available. The analysis of supersaturated designs rely on the assumption of effect sparsity (Box and Meyer, 1986). This assumes that only a few dominant factors actually affect the response. Satterthwaite (1959), proposed the idea of supersaturated design as a random balance design. Booth and Cox (1962), first examined two-level supersaturated designs ∗ Corresponding author. E-mail address: ckoukouv@math.ntua.gr (C. Koukouvinos). 0378-3758/$ - see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jspi.2004.11.002