An algorithm for constructing optimal resolvable incomplete block designs

This paper describes an efficient algorithm for the construction of optimal or near-optimal resolvable incomplete block designs (IBDs) for any number of treatments v < 100. The performance of this algorithm is evaluated against known lattice designs and the 414 or-designs of Patterson & Williams [36]. For the designs under study, it appears that our algorithm is about equally effective as the simulated annealing algorithm of Venables & Eccleston [42]. An example of the use of our algorithm to construct the row (or column) components of resolvable row-column designs is given.     

Error variance bias in neighbour balance and evenness of distribution designs

Neighbour balance and evenness of distribution designs help to address user concerns in the two‐dimensional layout of agricultural field trials. This is done by minimising the occurrence of pairwise treatment plot neighbours and ensuring that the replications of treatments are spread out across rows and columns of a trial. Such considerations result in a restriction on the normal randomisation process for a row‐column design which can lead to error variance bias. In this paper, uniformity trial data is used to assess the degree of this bias for both resolvable and non‐resolvable designs. Comparisons are made with a similar investigation using Linear Variance spatial designs.     

The Construction of Efficient Two-replicate Row–Column Designs for Use in Field Trials

Two-replicate row–column designs are often used for field trials in multisite tree or plant breeding programmes. With only two replicates for each trial, it is important to use designs with optimal or near optimal efficiency factors. This paper presents an algorithm for generating such designs. The method extends the contraction approach of Bailey and Patterson to any set of parameters and uses the factorial design construction algorithm of Williams and John to generate designs. Our experience with the algorithm is that it produces designs that are at least as good as, and often much better and more quickly generated than, those obtained by other recent computer algorithms.     

AN ALGORITHM FOR CONSTRUCTING OPTIMAL RESOLVABLE ROW-COLUMN DESIGNS

This paper describes an effective algorithm for constructing optimal or near-optimal resolvable row-column designs (RCDs) with up to 100 treatments. The performance of this algorithm is assessed against 20 2-replicate resolvable RCDs of Patterson & Robinson (1989) and 17 resolvable RCDs based on generalized cyclic designs (GCDs) of Ipinyomi & John (1985). The use of the algorithm to construct RCDs with contiguous replicates is discussed.     

ROW AND COLUMAN DESIGNS WITH CONTIGUOUS REPLICATES

A method is given for constructing row and column designs for situations where replicates are contiguous. Designs of this type are needed in cotton variety trials. A table of generating arrays is given from which a series of resolvable designs can be constructed; these designs are called latinized α-designs. Some results from cotton variety trials are presented.     

CONSTRUCTION OF RESOLVABLE ROW-COLUMN DESIGNS USING SIMULATED ANNEALING

The paper describes an algorithm which constructs optimal or near optimal resolvable row-column designs. Its performance is assessed against available tables of two-replicate designs.     

ADJUSTED ORTHOGONAL NESTED ROW-COLUMN DESIGNS

Adjusted orthogonality in nested row-column designs is defined and a sufficient condition established for its existence. It is shown that the properties of an adjusted orthogonal nested row-column design are directly related to those of its separate row and column component designs. A method for constructing efficient adjusted orthogonal designs involving a single replicate of every treatment in each of two blocks is given.     

Resolvable Designs With Unequal Block Sizes

Resolvable block designs for v varieties in blocks of size k require v to be a multiple of k so that all blocks are of the same size. If a factorization of v is not possible then a resolvable design with blocks of unequal size is necessary. Patterson & Williams (1976) suggested the use of designs derived from α -designs and conjectured that such designs are likely to be very efficient in the class of resolvable designs with block sizes k and k – 1. This paper examines these derived designs and compares them with designs generated directly using an interchange algorithm. It concludes that the derived designs should be used when v is large, but that for small v they can be relatively inefficient.     

A REVIEW OF BOUNDS FOR THE EFFICIENCY FACTOR OF BLOCK DESIGNS

This paper draws together bounds for the efficiency factor of block designs, starting with the papers of Conniffe & Stone (1974) and Williams & Patterson (1977). By extending the methods of Jarrett (1983), firstly to cover supercomplete block designs and then to cover resolvable designs, a set of bounds is obtained which provides the best current bounds for any block design with equal replication and equal block size, including resolvable designs and two-replicate resolvable designs as special cases. The bounds given for non-resolvable designs apply strictly only to designs which are either regular-graph (John & Mitchell, 1977) or whose duals are regular-graph. It is conjectured (John & Williams, 1982) that they are in fact global bounds. Similar qualifications apply to the bounds for resolvable designs.     

RANDOMIZED SEARCH STRATEGIES FOR FINDING OPTIMAL OR NEAR OPTIMAL BLOCK AND ROW-COLUMN DESIGNS

We consider a family of effective and efficient strategies for generating experimental designs of several types with high efficiency. These strategies employ randomized search directions and at some stages allow the possibility of taking steps in a direction of decreasing efficiency in an effort to avoid local optima. Hence our strategies have some affinity with the simulated annealing algorithm of combinatorial optimization. The methods work well and compare favourably with other search strategies. We have implemented them for incomplete block designs, optionally resolvable, and for row-column designs.     

αn–Designs

This paper defines a broad class of resolvable incomplete block designs called α_n–designs, of which the original α–designs are a special case with n = 1. The statistical and mathematical properties of α–designs extend naturally to these n –dimensional designs. They are a flexible class of resolvable designs appropriate for use in factorial experiments, in constructing efficient t –latinized resolvable block designs, and for enhancing the existing class of α–designs for a single treatment factor.     

t-Latinized Designs

A variety trial sometimes requires a resolvable block design in which the replicates are set out next to each other. The long blocks running through the replicates are then of interest. A t -latinized design is one in which groups of these t long blocks are binary. In this paper examples of such designs are given. It is shown that the algorithm described by John & Whitaker (1993) can be used to construct designs with high average efficiency factors. Upper bounds on these efficiency factors are also derived.     

A class of saturated row–column designs

A new class of row–column designs is proposed. These designs are saturated in terms of eliminating two-way heterogeneity with an additive model. The (m,s)-criterion is used to select optimal designs. It turns out that all (m,s)-optimal designs are binary. Square (m,s)-optimal designs are constructed and they are treatment-connected. Thus, all treatment contrasts are estimable regardless of the row and column effects.     

Optimality and Contrasts in Block Designs with Unequal Treatment Replication

The computer construction of optimal or near‐optimal experimental designs is common in practice. Search procedures are often based on the non‐zero eigenvalues of the information matrix of the design. Minimising the average of the pairwise treatment variances can also be used as a search criterion. For equal treatment replication these approaches are equivalent to maximising the harmonic mean of the design's canonical efficiency factors, but differ when treatments are unequally replicated. This paper investigates the extent of these differences and discusses some apparent inconsistencies previously observed when comparing the optimality of equally and unequally replicated designs.     

On the Construction of Nearest–Neighbour Balanced Row–Column Designs

Nearest–neighbour balance is considered a desirable property for an experiment to possess in situations where experimental units are influenced by their neighbours. This paper introduces a measure of the degree of nearest–neighbour balance of a design. The measure is used in an algorithm which generates nearest–neighbour balanced designs and is readily modified to obtain designs with various types of nearest–neighbour balance. Nearest–neighbour balanced designs are produced for a wide class of parameter settings, and in particular for those settings for which such designs cannot be found by existing direct combinatorial methods. In addition, designs with unequal row and column sizes, and designs with border plots are constructed using the approach presented here.     

A Note on the Moments and Computer Generation of the Shifted Gompertz Distribution

In this paper we give a class of row-column designs with the property that the i-th row and the j-th column have precisely r treatments in common. A conjecture that such designs are quasi-factorial is disproved by showing that the designs given in this paper are not quasi-factorial. It is also shown that the designs given here are nearly optimal.     

UPPER BOUNDS FOR LATINIZED DESIGNS

Upper bounds axe derived for the efficiency factor of a class of resolvable incomplete block designs known as latinized designs. These designs are particularly useful in glasshouse and field trials, and can be readily extended to two-dimensional blocking structures. Existing bounds for resolvable designs axe also reviewed and a comparison is made between the third moment bounds discussed by Jarrett (1989) and the second moment bounds of Tjur (1990).     

A NEW UPPER BOUND FOR THE EFFICIENCY FACTOR OF A BLOCK DESIGN

An upper bound for the efficiency factor of a block design, which in many cases is tighter than those reported by other authors, is derived. The bound is based on a lower bound for E(1/X) in terms of E(X) and var(X) for a random variable X on the unit interval. For the special case of resolvable designs, an improved bound is given which also competes with known bounds for resolvable designs in some cases.     

Balanced incomplete Latin square designs

Latin squares have been widely used to design an experiment where the blocking factors and treatment factors have the same number of levels. For some experiments, the size of blocks may be less than the number of treatments. Since not all the treatments can be compared within each block, a new class of designs called balanced incomplete Latin squares (BILS) is proposed. A general method for constructing BILS is proposed by an intelligent selection of certain cells from a complete Latin square via orthogonal Latin squares. The optimality of the proposed BILS designs is investigated. It is shown that the proposed transversal BILS designs are asymptotically optimal for all the row, column and treatment effects. The relative efficiencies of a delete-one-transversal BILS design with respect to the optimal designs for both cases are also derived; it is shown to be close to 100%, as the order becomes large.     

An algorithmic approach to construct D-optimal saturated designs for logistic model

In this paper, locally D-optimal saturated designs for a logistic model with one and two continuous input variables have been constructed by modifying the famous Fedorov exchange algorithm. A saturated design not only ensures the minimum number of runs in the design but also simplifies the row exchange computation. The basic idea is to exchange a design point with a point from the design space. The algorithm performs the best row exchange between design points and points form a candidate set representing the design space. Naturally, the resultant designs depend on the candidate set. For gain in precision, intuitively a candidate set with a larger number of points and the low discrepancy is desirable, but it increases the computational cost. Apart from the modification in row exchange computation, we propose implementing the algorithm in two stages. Initially, construct a design with a candidate set of affordable size and then later generate a new candidate set around the points of design searched in the former stage. In order to validate the optimality of constructed designs, we have used the general equivalence theorem. Algorithms for the construction of optimal designs have been implemented by developing suitable codes in R.