See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/334248944

Construction of Blocked Factorial Designs to Estimate Main Effects and

Selected Two-Factor Interactions

Preprint · July 2019

CITATIONS READS

0 131

1 author:

Janet Godolphin
University of Surrey
31 PUBLICATIONS 193 CITATIONS

SEE PROFILE

All content following this page was uploaded by Janet Godolphin on 19 August 2019.

The user has requested enhancement of the downloaded file.

 Construction of Blocked Factorial Designs to Estimate Main
Effects and Selected Two-Factor Interactions

J.D.Godolphin
Department of Mathematics, University of Surrey, UK, GU2 7XH.

Abstract Two-level factorial designs are widely used in industrial experiments. For processes
involving n factors, the construction of designs comprising 2n and 2n−p factorials, arranged in
blocks of size 2q is investigated. The aim is to estimate all main effects and a selected subset
arXiv:1907.02373v1 [math.ST] 4 Jul 2019

of two-factor interactions. Designs constructed according to minimum aberration criteria are

shown to not necessarily be the most appropriate designs in this situation. A design construc-
tion approach is proposed which exploits known results on proper vertex colourings in graph
theory. Examples are provided to illustrate the results and construction strategies. Particular
consideration is given to the special case of designs with blocks of size four and tables of designs
are given for this block size.

Keywords confounding, chromatic number, factorial effect, minimum aberration, resolution,

vertex colouring.

1 Introduction
In full and fractional factorial experiments, sources of systematic variation can be controlled
by grouping runs into blocks of 2q heterogeneous experimental units enabling the estimation of
effects with greater precision. For blocking to be effective, blocks should generally be relatively
small. Mead et al. (2012, Chapter 15) suggest that for many experiments, blocks should contain
no more than eight experimental units. However, the arrangement of a factorial experiment in
small blocks puts significant constraints on its estimability capability, due to the large number
of factorial effects confounded with blocks. Thus, the construction of such designs with desirable
properties can be challenging.
We use the notation 2n−p factorial for a regular fractional factorial in n two-level factors and
with 2n−p runs. A 2n−p factorial is completely determined by p independent defining words,
which generate the treatment defining contrast sub-group. The smallest number of factors in-
volved in a word in this sub-group is the resolution of the design. Estimability capability can
vary considerably between 2n−p factorials with the same resolution. The capacity of a design
to provide estimates can be quantified under the assumptions of the effect hierarchy principle,
namely:
H1: For m1 < m2 , m1 -factor interactions are more likely to be important than m2 -factor
interactions.
H2: For given m, all m-factor interactions are equally important.
See Wu and Hamada (2009, Chapter 4). Under assumptions H1 and H2, the minimum aberration
criterion, introduced by Fries and Hunter (1980), ranks 2n−p factorials by the distribution of word
lengths in the treatment defining contrast sub-group. Tables of designs according to this criterion
are provided by Chen et al. (1993) and Block and Mee (2005). Several papers, including Hedayat

1
and Pesotan (1992), Wu and Chen (1992) and Wang (2007), address the problem of selecting 2n−p
factorials in situations where assumption H2 does not hold in that estimates of main effects and
selected two-factor interactions are required. Wu et al. (2012) give tables of resolution IV 2n−p
factorials for up to 128 runs which cover all estimable configurations of two-factor interactions.
When arranging 2n factorials in blocks, two assumptions are made:
B1: Interactions between block factors and treatment factors are negligible.

B2: An interaction between two or more block factors has the same importance as the main
effect of a block factor.
The blocking framework for a 2n factorial arranged in blocks of size 2q is typically described by
means of a blocking defining contrast sub-group. This has the same mathematical structure as
the treatment defining contrast sub-group for a 2n−(n−q) factorial, and is determined by n − q
independent defining words. Sun et al. (1997) apply the minimum aberration criterion to the
block defining contrast sub-group and give optimal blocking schemes for n ≤ 8.
The problem of selecting blocked 2n−p factorials with good estimability capability is con-
siderably more complex than that of selecting a 2n−p factorial or of blocking a 2n factorial,
being complicated by the need to take account of both treatment and block defining contrast
sub-groups. A substantial body of work focuses on construction of optimal blocked 2n−p fac-
torials. Sun et al. (1997) and Mukerjee and Wu (1999) use admissibility criteria based on both
word length distributions. Other work involves various methods of combining the word length
distributions of the two defining contrast subgroups into a single distribution, and using the
minimum aberration criterion on this combined distribution. See for example Sitter et al. (1997),
Chen and Cheng (1999), Zhang and Park (2000), Cheng and Wu (2002), Xu and Lau (2005)
and Xu and Mee (2010). Instead of employing a word length approach, Cheng and Mukerjee
(2001) use a criterion appertaining to the alias pattern of the interactions. Several of the papers
referred to above include extensive tables of designs. In particular, Sun et al. (1997) constitutes
a comprehensive design catalogue for n ≤ 9 and up to 128 runs.
Prior knowledge about the process under investigation can result in a subset of the two-factor
interactions being of particular interest. The aim of this work is to develop a strategy for the
bespoke arrangement of 2n and 2n−p factorials in blocks of size 2q so that all main effects and
selected two-factor interactions can be estimated. We make assumptions H1, B1 and B2, but
digress from the effect hierarchy principle in that some two-factor interactions are considered
more important than others. In many cases a blocked full or fractional factorial will not provide
estimates of all main effects and all two-factor interactions. A design approach which prioritises
estimation of main effects and has practitioner led focus on those two-factor interactions thought
most likely to be relevant therefore seems pragmatic. No two-factor interactions are assumed
negligible: the estimation of any of these in addition to the specified ones is advantageous.
Interactions in three or more factors are assumed to be negligible. Henceforth, unless otherwise
stated, interaction will refer to a two-factor interaction.
Designs in the literature comprising 2n and 2n−p factorials in blocks of size 2q yielding
estimates of all main effects and some interactions are generally chosen to maximise the number
of estimable interactions, under assumption H2. If a specific set of interactions are of interest,
it may not be possible to find a design in the literature which accommodates all main effects
and all chosen interactions. This can arise for two reasons. First, it might be impossible to
construct a design providing the desired estimates. This depends on both the number and
configuration of the selected interactions. Second, it may be possible to construct a design which
accommodates the required estimates, but if such a design does not rank highly under optimality
criteria then it will not appear in design tables. Further, even if a selected set of interactions can

2
be accommodated by a listed design, the problem of mapping the factors of an experiment onto
the factors of a design so all chosen interactions are estimable is not necessarily trivial.
In this work, instead of using the block defining contrast sub-group, we identify the blocking
structure of a design by the sub-group of treatment combinations comprising the principal block.
With blocks of size 2q , this requires q independent treatment combinations, rather than n − q or
n − p − q independent defining words, which is advantageous when q is small compared to n − q
or n − p − q. Further, these independent treatment combinations correspond to a q × n generator
matrix which gives valuable and immediate information on the estimability of main effects and
interactions.
In §2, fundamental concepts on generator matrices are introduced and a graph theory ap-
proach is developed for the construction of designs comprising 2n factorials in blocks of size 2q .
By representing the selected interactions graphically, known results in graph theory yield condi-
tions on the interaction set which guarantee the existence of a design that provides the required
estimates. For interaction sets satisfying the conditions, the construction problem is expressed as
a vertex colouring task in graph theory. A valid colouring leads to a factor grouping associated
with a profile set, which in turn yields a generator matrix and hence a suitable design. Thus,
rather than seeking an existing design to use for an experiment, a design is constructed which
is tailored to the set of chosen interactions. Graph theory terminology is consistent with Bondy
and Murty (2008). In §3, the method is extended to encompass constructions of blocked 2n−p
factorials. In §4 we focus on the special case of designs with blocks of size four. This constitutes
an important design class since the blocks are small enough for blocking to be effective in most
cases, but large enough to allow the estimation of all main effects and some two-factor interac-
tions. Design templates are provided for 2n−p factorials in blocks of size four for up to twelve
factors and 128 runs.

2 Blocked full factorial designs

2.1 Preliminaries
Treatment combinations are represented in two equivalent forms: in lower case letters and as
row vectors (x1 , x2 , · · · , xn ) in the Galois Field of order 2, GF(2), with xj = 1 if the jth factor
is at high level, and xj = 0 otherwise. The treatment combination with all factors at low level
is denoted by (1) or by the 1 × n vector with all terms zero. Factors are labelled F1 , · · · , Fn , or
alphabetically A, B, . . . in examples. Thus, in a five factor process, the treatment combination
with A, C, D high and B, E low is represented as acd and (1, 0, 1, 1, 0).
A blocked 2n factorial has the treatment combinations arranged in 2n−q blocks of size 2q . The
blocking structure can be identified by q treatment combinations represented by q independent
row vectors denoted xi = (xi1 , xi2 , · · · , xP in ), for i = 1, . . . , q. The principal block contains the
q
2q treatment combinations of the form i=1 αi xi , where αi ∈ {0, 1} and addition is modulo
2. These form an Abelian sub-group of the 2n treatment combinations and remaining blocks
contain cosets of this sub-group. Thus the design is fully determined by the q × n generator
matrix of rank q:    
x1 x11 x12 · · · x1n
X =  ...  =  ... .. ..  .
  
. . 
xq xq1 xq2 · · · xqn
Example 1: Consider design D1, comprising a 25 factorial arranged in blocks of size four with

3
generator matrix:  
1 1 1 0 0
X1 = .
1 0 1 1 1
The principal block of D1 contains (1), abc, acde and bde. The design is given below, with
columns representing blocks.

(1) a b c d e ad ae
abc bc ac ab abcd abce bcd bce
. (2.1)
acde cde abcde ade ace acd ce cd
bde abde de bcde be bd abe abd

In a blocked 2n factorial, the 2n − 1 factorial effects separate into two sets: 2n−q − 1 effects
which are confounded with blocks, and 2n − 2n−q estimable effects. Estimability of main effects
and interactions can be deduced directly from the generator matrix.

Theorem 1 A factorial effect is estimable from a blocked 2n factorial iff at least one of the
generating treatment combinations contains an odd number of effect terms.

Proof: Consider any factorial effect. Treatment combinations with an odd number of terms
in common with the effect have one sign in the effect contrast, and treatment combinations
with an even number of terms in common with the effect have the opposite sign. Thus, for the
effect of the principal block to sum out in the linear combination of observations relating to the
factorial effect, 2q−1 treatment combinations of the principal block must have an odd number of
effect terms and 2q−1 must have an even number. This occurs iff at least one of the generating
treatment combinations contains an odd number of effect terms. By the method of construction,
every block contains 2q−1 treatment combinations of each sign in the effect contrast, enabling
all block effects to be eliminated, iff this property is exhibited by the principal block. This
establishes the result.
Theorem 1 leads to results on estimability of main effects and interactions in terms of prop-
erties of the design generator matrix. To state the results efficiently, we introduce the notation
Xu to denote the set of 2u − 1 non-zero u × 1 vectors over GF(2). For example:
   
1 1 0
X2 = , , .
1 0 1

From Theorem 1, a necessary and sufficient condition for a main effect, Fj say, to be estimable
is that at least one of xij is non-zero, for i = 1, . . . , q. Thus,
Corollary 1 All main effects are estimable from a blocked 2n factorial iff every column of X is
a column of Xq .

In this work, only designs from which all main effects are estimable will be considered, that
is, all blocked 2n factorials are required to satisfy Corollary 1 and therefore to have generator
matrices consisting of columns of Xq . Further, since generator matrices are full rank, every
generator matrix must contain at least q linearly independent vectors of Xq . Now consider an
interaction, Fj Fk say. By Theorem 1, this is estimable iff there exists some i ∈ {1, . . . , q} for
which exactly one of xij and xik is non-zero. This gives:
Corollary 2 An interaction is estimable from a blocked 2n factorial iff the corresponding two
columns of X are different.

4
Example 1 continued: By Corollary 1, all five main effects are estimable from D1. By Corol-
lary 2, D1 provides estimates of eight of the ten interactions. The first and third columns of
X1 are the same indicating that AC is not estimable. Similarly, DE is not estimable. These
properties are confirmed by inspection of (2.1).
Corollaries 1 and 2 give an indication of the limitations of the capacity of a blocked 2n
factorial to estimate interactions. From Corollary 1, there are 2q − 1 different columns available
for inclusion in X. From Corollary 2 an interaction is only estimable if the corresponding columns
of X are different. Note that for a design in blocks of size two, that is with q = 1, the implications
of Corollaries 1 and 2 are that if all main effects are estimable then no two-factor interactions
are estimable, since X1 only contains one vector. In general, if n > 2q − 1 not all interactions
will be estimable. The number of factors in the design can be expressed as n = (2q − 1)v + w,
where v = [n/(2q − 1)], with [.] denoting the integer part of, and w ∈ {0, 1, . . . , 2q − 2}.
Theorem 2 An upper bound for the number of estimable interactions from a blocked 2n factorial
is:    
n q v
φmax = − vw − (2 − 1) . (2.2)
2 2
Proof: A blocked 2n factorial has generator matrix X with columns comprising members of Xq .
From Corollary 2, the number of estimable interactions is the number of pairs of columns of X
in which the columns are different. Expressing the number of columns of X as n = (2q − 1)v + w,
an upper bound for the number of pairs of distinct columns in X is:
     
n v+1 q v+1
−w − (2 − 1 − w) .
2 2 2
After manipulation of the Binomial coefficients involving v + 1, this expression gives the right
hand side of (2.2) and the result follows.
A similar result is given in Godolphin (2019) in relation to factorial designs arranged in a
rectangular array. Theorem 2 prompts:
Corollary 3 If n ≤ 2q − 1 then all interactions in a blocked 2n factorial are estimable, provided
that X is formed from n distinct members of Xq and has rank q, over GF(2). If n > 2q − 1 the
number of estimable interactions in a blocked 2n factorial achieves φmax iff X is formed from
v + 1 copies of each of w columns of Xq and of v copies of each of the remaining 2q − 1 − w
columns.
Corallary 3 is illustrated by the following examples:
Example 2: A 26 factorial is to be arranged in blocks of size eight. Since n = 6 < 7 = 2q − 1, all
main effects and interactions are estimable from a design with generator matrix having columns
comprising six distinct members of X3 . Note that any six members of X3 yield a full rank
generator matrix. Such a matrix is
 
1 0 0 1 1 0
X2 =  0 1 0 1 0 1  ,
0 0 1 0 1 1
giving design, D2 say, with principal block containing the treatment combination subgroup (1), ade, bdf,
abef, cef, acdf, bcde and abc. The seven remaining blocks contain cosets.
Example 3: For a 26 factorial in blocks of size four, φmax = 12. This is achieved by D3 with a
generator matrix formed from two copies of each column of X2 :
 
1 1 0 0 1 1
X3 = .
0 0 1 1 1 1

5
The principal block of D3 contains (1), abef, cdef and abcd, and the 15 remaining blocks contain
cosets. Design D3 provides estimates of all interactions except AB, CD and EF .
There are situations in which a design providing estimates of fewer than φmax interactions is
preferable to one achieving the bound.
Example 4: A 26 factorial in blocks of size four is required to assess all main effects and all
interactions involving one factor, A say. Use of D3 is unsatisfactory since, even with a reordering
of the columns of X3 , one of the interactions of interest will be inestimable. An alternative design,
D4, is suggested, with generator matrix:
 
1 0 0 0 1 1
X4 = .
0 1 1 1 1 1
Design D4 gives estimates of 11 interactions, including all those involving A.

2.2 Graphical representation of required and estimable interactions

When planning an experiment it is informative to represent interactions of interest by means
of a labelled graph with vertices labelled to correspond to factors and with an edge between
vertices Fi and Fj if interaction Fi Fj is of interest. Hedayat and Pesotan (1992), Wu and Chen
(1992) and Wu et al. (2012) use such graphs to aid in the construction of designs to estimate
main effects and selected interactions in experiments without blocking. Wu and Chen (1992)
coin the phrase requirements graph which we adopt in this work. Requirements graphs have the
advantages of giving a visual depiction of the way in which interactions of interest are interlinked
and of making results in graph theory available as tools in the design planning process. Figure
1(i) depicts the requirements graph for the problem of Example 4. The graph of estimable
interactions for a design will be termed the estimability graph. Figure 1(ii) gives the estimability
graph for D4. This contains the five edges of the requirements graph and six additional edges,
indicating that BE, BF , CE, CF , DE and DF are estimable as well as the selected interactions.
The estimability graph is a graphical representation of X4 since each edge coincides with a
pair of generator matrix columns that are not equal. Note that the estimability graph has the
requirements graph as a sub-graph.
The estimability graph for D4, is a labelled complete 3-partite graph. In general, a labelled
complete r-partite graph on n vertices has the n labelled vertices assigned to r non-empty disjoint
sets. There is an edge between every pair of vertices from different sets and there are no edges
between vertices from the same set. In D4 the three vertex sets are {B, C, D}, {E, F } and {A}.
We denote the set of labelled complete r-partite graphs on n vertices, for r taking all values
in {q, q + 1, . . . , 2q − 1}, as T (n, q). Every design comprising a 2n factorial in blocks of size 2q
corresponds to a unique estimability graph in T (n, q). For example, the estimability graph for
D4, given in Figure 1(ii), is a member of T (6, 2).

2.3 Equivalent designs

Consider X4 , the generator matrix for D4. This contains columns of X2 with multiplicities one,
two and three. Any other 26 factorial arranged in blocks of size four which also has a generator
matrix comprising columns of X2 with multiplicities three, two and one will have an estimability
graph in T (6, 2) which is isomorphic to (or possibly the same as) the estimability graph for D4
and is said to be equivalent to D4. In general, two designs which have isomorphic estimability
graphs in T (n, q) are equivalent in that they provide estimates of equivalent sets of two factor
interactions. Note however that for q > 2 the designs themselves are not necessarily isomorphic:
for example they may provide estimates of different numbers of three factor interactions.

6
 Figure 1: Graphs for Example 4.
B B

C E

A D A C

E F

F D

(i) D4 requirements graph (ii) D4 estimability graph

Figure 2: Graphs for Example 4 with vertex colouring.

B1 B1

C1 E2

A3 D1 A3 C1

E2 F2

F2 D1

(i) D4 requirements graph (ii) D4 estimability graph

7
 A design, D, comprising a 2n factorial in blocks of size 2q has generator matrix, X, containing
at least q independent members of Xq . Let the multiplicities of the members of Xq in X be
P2q −1
n1 ≥ n2 ≥ · · · ≥ n2q −1 ≥ 0, where nq > 0 and i=1 ni = n. Then any other blocked 2n factorial
having a full rank q × n generator matrix with the same multiplicities is equivalent to D. Thus,
subject to the requirement that the generator matrix has rank q, the estimability properties
of a blocked 2n factorial are wholly determined by the multiplicities of the generator matrix
columns. This prompts efficient representation of classes of designs with equivalent properties
and of individual designs:
Definition: A 2n factorial in blocks of size 2q is said to have profile set hn1 , n2 , . . . , n2q −1 i where
n1 ≥ n2 ≥ · · · ≥ n2q −1 ≥ 0 are the multiplicities of columns of Xq in the generator matrix X.
Definition: Consider a 2n factorial in blocks of size 2q with profile set hn1 , n2 , . . . , n2q −1 i. Let
q0 be the largest value such that nq0 > 0. For i ≤ q0 denote the factors relating to entry ni in
the profile set by Fi,1 , . . . , Fi,ni . Then the design and its properties are determined by the factor
grouping {(F1,1 , . . . , F1,n1 ), · · · , (Fq0 ,1 , . . . , Fq0 ,nq0 )}.
Two designs with the same profile set have isomorphic estimability graphs and are equivalent.
The set of estimable interactions from a design are obtained directly from the factor grouping.
This is illustrated by reference to D3 and D4. Design D4 has profile set h3, 2, 1i and any 26
factorial in blocks of size four with generator matrix comprising columns of X2 with multiplicities
one, two and three will have the same profile set and will be equivalent to D4. The estimability
properties of D4 are specified by the factor grouping {(B, C, D), (E, F ), (A)}: an interaction is
estimable iff the factors are in different sets of the factor grouping: for example, BE is estimable
but BC is not estimable. By comparison, D3, with profile set h2, 2, 2i and factor grouping
{(A, B), (C, D), (E, F )}, is not equivalent to D4. Designs D3 and D4 are equivalent to designs
labelled 6-01/B4.1 and 6-01/B4.2, considered in Case 1 of Sun et al. (1997), where they are
specified in terms of four effects which generate the block defining contrast subgroup. (In fact,
since q = 2, D3 is isomorphic to 6-01/B4.1 and D4 is isomorphic to 6-01/B4.2). The design
summaries in Sun et al. (1997) indicate that all main effects are estimable from both designs,
and give the number of estimable interactions. In contrast with the generator matrix approach,
the estimable interactions for each design are not evident without some work. Under H2 that
effects of the same order are equally important, based on the number of estimable interactions
6-01/B4.1 (D4) would be considered inferior to 6-01/B4.2 (D3), which has minimum aberration
blocking scheme. However, as demonstrated in Example 4, when H2 does not hold because a
subset of interactions are of particular interest, there are situations in which 6-01/B4.1 (D4)
would be preferred over 6-01/B4.2 (D3).

2.4 Design construction via vertex colouring

In Figure 2, colours 1, 2, 3 are applied to vertices of the D4 requirements and estimability graphs,
with vertices in the factor grouping set corresponding to ni assigned colour i. For both graphs
this results in a vertex colouring such that adjacent vertices, that is vertices which are joined
by an edge, are coloured differently. In graph theory terms, the vertex colourings in Figure 2
constitute proper 3-colourings and demonstrate that the graphs are k-chromatic, for some value
k ≤ 3. See Bondy and Murty (2008, Chapter 14). If Example 4 was amended so that BD, BE
and DE were required as well as interactions involving A, this would give a new requirements
graph with eight edges and, due to the arrangement of the edges, it would not be possible to
assign colours 1, 2, 3 to the vertices to give a proper 3-colouring. Thus, there is no graph in
T (6, 2) with the new requirements graph as a subgraph and no 26 factorial in blocks of size four
gives estimates of all main effects and the eight interactions. The general point is summarised

8
as follows:
Lemma 1 The n main effects and the interactions of a requirements graph can be accommodated
by a 2n factorial in blocks of size 2q iff the requirements graph is k-chromatic for some value
k ≤ 2q − 1.
Lemma 1 prompts the strategy that will be adopted for design construction. A set of interactions
is selected such that the requirements graph is k-chromatic with k ≤ 2q − 1. Let k∗ = min{2q −
1, n}. A colouring in k0 colours is identified for some k0 such that min{q, k} ≤ k0 ≤ k∗ . The
k0 -colouring corresponds to an estimability graph in T (n, q), with the requirements graph as
a sub-graph, and to a factor grouping. The practitioner uses the k0 -colouring to identify a
generator matrix for a 2n factorial in blocks of size 2q that yields all the required estimates. The
factor grouping gives a concise means of recording the estimable interactions.
Note that if n ≤ 2q − 1 then each vertex can be allocated a unique colour, to give an n-
colouring. Any full rank matrix with columns comprising n members of Xq will be a suitable
generator matrix. The generator matrix X2 of Example 2 has this form and the corresponding
estimability graph is a complete 6-partite graph in T (6, 3). If n > 2q − 1 then not all interactions
can be estimated. Hence, there are sets of interactions which yield a requirements graph that
is k-chromatic with k > 2q − 1. In such cases, no valid colouring is possible in 2q − 1 colours
or fewer. Thus, the requirements graph will not be a subgraph of any graph in T (n, q) and no
blocked 2n factorial will provide all the desired estimates. Known results in graph theory are
useful in providing guidance on interaction sets that can be accommodated by designs.

Theorem 3 Let S denote a subset of interactions, Fi Fj , for an n factor process. If one or both
of the following conditions are satisfied, then a 2n factorial in blocks of size 2q exists from which
all main effects and all interactions in S can be estimated:
(i) S contains no more than 2q − 1 interactions relating to any factor and there is no set of
2q factors for which S contains all q2 pairwise interactions;

(ii) There are at most 2q − 1 factors for which S contains at least 2q − 1 interactions.

Proof: Let G be the requirements graph with n vertices and with edges relating to interactions
in S. By Lemma 1, a 2n design in blocks of size 2q can be obtained with the required properties
iff G is k-chromatic for some k ≤ 2q − 1. The two cases are considered separately. If S satisfies
(i), then the maximum vertex degree of G does not exceed 2q − 1 and G does not contain the
complete graph on 2q vertices. By Brooks’ Theorem (see Brooks (1941)), G is k-chromatic for
some k ≤ 2q − 1. If S satisfies (ii), then G has fewer than 2q vertices of degree at least 2q − 1.
By Theorem 14.6 of Bondy and Murty (2008), every k-chromatic graph has at least k vertices of
degree at least k − 1. Thus G is k-chromatic for some k ≤ 2q − 1. The result follows.
When planning an experiment, if the practitioner lists interactions in order of priority then
Theorem 3 will give guidance on the selection of an interaction set which gives rise to a require-
ments graph with chromatic number no greater than 2q − 1. Obtaining a valid colouring with
no more than 2q − 1 colours for such a graph is generally straightforward. However, if it proves
challenging either to determine the chromatic number for a given requirements graph, or to ob-
tain a valid colouring with no more than 2q − 1 colours, then graph theory routines in packages
such as the Sage computer algebra system (see Stein et al. (2018)) are freely available.
Example 5 (part 1): Two designs comprising 27 factorials in blocks of size four are sought,
each with a different set of ten interactions of interest:
S1 = {AB, AC, AD, BC, BE, CD, DF, EF, EG, F G};

9
 Figure 3: Graphs for DS1 of Example 5 (part 1) with interaction set S1
B1 E2 B1
C3
A2

A2 C3 G1 D1

E2
F3
D1 F3 G1
Requirements graph Estimability graph

Figure 4: Graphs for DS2 of Example 5 (part 1) with interaction set S2

A1
B2 E3
B2 C3
D1
F1
A1 G2 D1

G2 E3
C3
F1
Requirements graph Estimability graph

S2 = {AB, AC, BC, BD, BE, CD, CF, CG, EF, EG}.
Interaction sets S1 and S2 each satisfy exactly one of the conditions of Theorem 3, with S1
satisfying (i) and S2 satisfying (ii). The requirements graphs, complete with valid 3-colourings,
and the corresponding estimability graphs are presented in Figures 3 and 4. Let DS1 and DS2
denote designs that coincide with the graph colourings of Figures 3 and 4 and which accommodate
S1 and S2 . In both cases the factor grouping, and hence a generator matrix, can be written down
from the requirements graph colouring: for example the factor grouping for DS1 consistent with
the 3-colouring in Figure 3 (i) is {(B, D, G), (A, E), (C, F )} and a generator matrix for DS1 is:
 
1 1 0 1 1 0 1
X= .
0 1 1 1 0 1 1

Here, the colours 1, 2, 3 have been associated with columns of X2 in the order (1, 1)T , (1, 0)T ,
(0, 1)T , but any one-to-one mapping between the colours (or factor sets) and the columns of X2
is equivalent. Likewise, DS2 has factor grouping {(A, D, F ), (B, G), (C, E)}. Designs DS1 and
DS2 are equivalent: both have profile sets h3, 2, 2i and isomorphic estimability graphs in T (7, 2)
and give estimates of 16 interactions.
Example 5 is returned to in subsequent sections.

2.5 Alternative colourings and profile sets

A graph which is k-chromatic may admit multiple proper k0 -colourings, for some k0 ≥ k, and
these will not necessarily share the same set cardinalities. Thus, a requirements graph with
k ≤ k∗ may permit different k0 -colourings, for some k0 with max{k, q} ≤ k0 ≤ k∗ , leading to
blocked 2n factorials with different profile sets and therefore able to estimate different numbers
of interactions. Further, for a requirements graph that is k-chromatic with k < k∗ , a colouring

10
 Figure 5: Graphs for DS2A of Example 5 (part 2) with interaction set S2
A1 B2
B2 E1
D1 F2
D1
F2
A1 G2 E1 G2

C3
C3
Requirements graph
Estimability graph

in k∗ colours can always be found. These points are now investigated briefly in the context of
constructing designs with desirable properties, the assumption being that if two alternative 2n
factorials in blocks of size 2q with different profile sets both give the required estimates, then a
design with profile set resulting in a larger number of estimable interactions is preferred. The
first point is now demonstrated.
Example 5 (part 2): The 3-colouring of the requirements graph for S2 given in Figure 4
yields DS2 with profile set h3, 2, 2i. In Figure 5, the S2 requirements graph is reproduced with
an alternative 3-colouring. This gives design DS2A with profile set h3, 3, 1i. The estimability
graph for DS2A is included in Figure 5. Of DS2 and DS2A, the former is preferred since it gives
estimates of 16 interactions compared to the 15 interactions estimable from DS2A. A 3-colouring
of the S2 requirements graph can also be given leading to a design with profile set h4, 2, 1i, yielding
estimates of only 14 interactions.
To investigate the second point, consider a k-chromatic requirements graph, where k < k∗ ,
together with a k0 -colouring for some k0 such that k0 < k∗ . If k0 ≥ q this colouring corresponds
to a design with profile set having n1 > 1, ni ≥ 1, for i = 2, . . . , k0 , and ni = 0, for i =
k0 + 1, . . . , 2q − 1. A blocked 2n factorial with full rank generator matrix comprising copies of k0
columns of Xq , arranged according to the profile set provides the required estimates. The graph
colouring can be adjusted to incorporate additional colours. To demonstrate this, arbitrarily
select [n1 /2] vertices of colour 1 and recolour these with colour k0 + 1. The new colouring is a
valid (k0 + 1)-colouring which leads to an estimability graph in T (n, q) and a blocked 2n factorial
giving estimates of an additional [n1 /2](n1 − [n1 /2]) interactions over those achieved by the
original design. The process can be repeated until a colouring is achieved with k∗ colours. If
k0 < q, the initial k0 -colouring will not correspond to a blocked 2n factorial, since the colouring
does not yield a full rank generator matrix. Use of the process as described for k0 ≥ q will result
in graph colourings using progressively more colours. Once a q-colouring is achieved, the graph
corresponds to a blocked 2n factorial and, after successively including more colours, ultimately
a k∗ -colouring is achieved.
In terms of the profile set, the number of estimable interactions from a blocked 2n factorial
P2q −2 P2q −1
is i=1 j=i+1 ni nj , which achieves the bound φmax of Theorem 2 when n1 − n2q −1 ≤ 1. In
the notation of Theorem 2 this occurs iff both n1 ≤ v + 1 and n2q −1 = v. Such a colouring
is termed an equitable k∗ -colouring. If n ≤ 2q − 1 an equitable n-colouring has each vertex
coloured uniquely: design D2 of Example 2, with profile set h1, 1, 1, 1, 1, 1, 0i, is such a case. The
colourings in Figure 3 and 4, leading to DS1 and DS2 are equitable k∗ -colourings, with k∗ = 3.
For n > 2q − 1, not all k-chromatic graphs with k ≤ 2q − 1 admit equitable k∗ -colourings: the
requirements graph of D3, displayed in Figure 1(i) cannot be equitably 3-coloured.

11
3 Blocked 2n−p factorial designs
Unless the number of factors is small it is unlikely that resources will be available to conduct
a full blocked 2n factorial. Instead, a 2n−p factorial is chosen, where p > 0, and the treatment
combinations of the principal fraction are arranged in blocks of size 2q according to a generating
matrix to form a blocked 2n−p factorial. A 2n−p factorial is specified by p independent defining
z zi,n−p
words F1 i,1 · · · Fn−p Fi for i = n − p + 1, . . . , n, with zi,j ∈ {0, 1}, summarised by the p × (n − p)
matrix Z = (zi,j ). The defining words give rise to a treatment defining contrast sub-group
containing 2p − 1 effects. The factorial effects not in the treatment defining contrast sub-group
are partitioned into 2n−p − 1 alias sets, each of size 2p .
The complexity of planning an experiment to estimate all main effects and selected interac-
tions is far greater when using a blocked 2n−p factorial than when using a blocked 2n factorial.
This is because consideration needs to be given to both the principal block generating sub-group
and the treatment defining contrast sub-group. The major difference between the constructions
of blocked 2n factorials and blocked 2n−p factorials proposed here is that, for the former the prac-
titioner is able to select all n columns of X whereas, for the latter, only the first n − p columns
are selected. A blocked 2n−p factorial has generator matrix X = (X I X II ). The columns of
the q × (n − p) sub-matrix X I are selected from Xq and the q × p sub-matrix X II is then de-
termined from X II = X I Z T , where calculations are modulo 2. In a 2n−p factorial arranged in
blocks of size 2q , of the alias sets, 2n−p−q − 1 are confounded with blocks and the remaining
2n−p − 2n−p−q are orthogonal to blocks. Whether a factorial effect is confounded with blocks or
not is determined directly from X as in §2. Theorem 1 is adjusted to include fractional designs:
Theorem 4 For a 2n−p factorial in blocks of size 2q , an effect is not confounded with blocks iff
at least one of the generating treatment combinations contains an odd number of effect terms.
Likewise, for effects of interest:
Corollary 4 For a 2n−p factorial in blocks of size 2q , all main effects are unconfounded with
blocks iff every column of X is a column of Xq .

Corollary 5 For a 2n−p factorial in blocks of size 2q , an interaction is not confounded with
blocks iff the corresponding two columns of X are different.

An effect of interest will be estimable iff both conditions are satisfied:

C1: all 2p − 1 aliases of the effect can be considered negligible and can be suppressed;
C2: the effect is not confounded with blocks.
Since all main effects are required and no two factor interactions are assumed negligible, only
2n−p factorials of resolution at least IV are used. Similarly, only selections of columns from Xq
for X I which yield X II sub-matrices also with columns all from Xq , that is with no zero sum
columns, are entertained. The design construction approach is illustrated:
Example 6: A design comprising a 26−1 factorial arranged in blocks of size eight is required, to
provide estimates of all main effects and as many interactions as possible.
A sensible starting point is to use a fraction with defining word of length five or six to yield a
26−1 factorial for which main effects and two factor interactions are only aliased with negligible
effects. Consider the defining word ABCDEF , giving Z = (1, 1, 1, 1, 1). We seek a 3 × 6
generator matrix X = (X I X II ) by selecting five of the seven members of X3 for the columns of
X I and then obtaining X II using X II = X I Z T . However, any choice of five distinct members of

12
X3 for X I yields a vector X II which is one of the selected members. Thus the generator matrix
contains a repeated column. For example:
 
1 0 0 1 1 1
X= 0 1 1 1 1 0 .
0 0 1 0 1 0
With this generator matrix, AF is confounded with blocks and so is not estimable, but all main
effects and the other 14 interactions are estimable. Clearly the design can be adjusted so that the
interaction thought to be of least interest is non-estimable.
The same duplicated column property is observed if a defining word of length five, such as
ABCDF , is used. A defining word of length four and careful choice of members of X3 for X I
can yield a matrix X with six distinct columns. However, although the resulting design will
have no main effects or two factor interactions confounded with blocks, three pairs of two factor
interactions will be aliased and so fail C2.
Example 7: An experiment involving an industrial process is planned in which a 29−2 factorial,
in factors A, B, . . . , I, is arranged in blocks of size eight. Due to their roles in the process, it is
thought that G, H and I do not interact with each other, but interactions between these factors
and the other six factors are of interest. Thus a design with profile set h3, 1, 1, 1, 1, 1, 1i, with
G, H and I forming the factor group of size three, would provide the required estimates and also
estimates of interactions between A, . . . , F . Use of the defining words ABEGH and ABCDEF I
gives a resolution V fraction and has
 
1 1 0 0 1 0 1
Z= .
1 1 1 1 1 1 0
Then, trial and error in selecting and arranging six different columns of X3 in X I yields the
generator matrix:  
1 0 0 0 1 1 1 1 1
X= 0 1 0 1 1 0 1 1 1 ,
0 0 1 1 0 1 1 1 1
which provides the estimates summarised above.

4 Blocks of size four

In this section we focus on the construction of 2n and 2n−p designs in blocks of size four. For
a factorial experiment arranged in blocks of size two, a design which provides estimates of all
main effects will not enable estimation of any interactions. Thus, the factorial experiments in
blocks of size four is the class with smallest block size enabling estimation of all main effects
and some interactions. Further, the blocks are small enough for blocking to be effective in most
cases. With the large number of factorial effects confounded with blocks, designs in blocks of
size four with desirable properties are challenging to construct. The importance of these designs
and the challenge of their construction merit the special attention given to the design class. To
arrange a 2n factorial in blocks of size four we use the approach of §2.4 to obtain information
on properties of interaction sets which give rise to requirements graphs with chromatic number
2 or 3. In the following result, conditions (i) and (ii) are covered by Theorem 3 but are included
here for completeness.
Theorem 5 Let S denote a subset of interactions, Fi Fj , for an n factor process. If one or more
of the following are satisfied, then a 2n factorial in blocks of size four exists from which all main
effects and all interactions in S can be estimated:

13
 (i) S contains no more than three interactions relating to any factor and there is no set of
four factors for which S contains all six pairwise interactions;
(ii) There are at most three factors for which S contains more than two interactions;
(iii) There is no subset of factors, Fi , for i = 1, . . . , m ≥ 3, for which all m interactions
F1 F2 , F2 F3 , . . . Fm−1 Fm , Fm F1 ∈ S, or there is at least one factor that is common to all
such subsets.
Proof of (iii): Let G be the requirements graph with n vertices and with edges relating to
interactions in S. By Lemma 1, a 2n design in blocks of size four can be obtained with the
required properties iff G is k-chromatic for k ≤ 3.
If S satisfies (iii) by there being no subset of factors, F1 , . . . , Fm , for m ≥ 3, such that
F1 F2 , F2 F3 , . . . Fm−1 Fm , Fm F1 ∈ S then G contains no cycles, that is G is acyclic, and is therefore
2-chromatic, (see Bondy and Murty (2008, Chapters 4, 14)). Suppose instead that S satisfies (iii)
by there being at least one subset of factors, F1 , . . . , Fm , for m ≥ 3, forming a cycle in G, with all
such subsets having at least one common factor. Without loss of generality let F1 be a common
factor for all subsets. Removal of F1 and adjacent edges from G gives a graph G ′ in n − 1 vertices
which is acyclic and is therefore 2-chromatic. Consider any 2-colouring of the vertices of G ′ and
reinstate F1 and incident edges to reform G. Vertex F1 can be coloured with a third colour to
give a valid 3-colouring of G. Hence G is k-chromatic, with k ≤ 3.
Use of Theorem 5 is demonstrated by the following continuation of Example 5:
Example 5 (part 3): Designs comprising 27 replicates in blocks of size four are sought, with
the following interaction sets of interest:
S3 = {AB, AD, AF, AG, BC, BD, CD, CE, DE, DF, DG};
S4 = {AB, AC, AD, AE, AG, BF, CD, CG, DG, EF }.
Set S3 satisfies condition (iii) of Theorem 5. The requirements graph, complete with valid
3-colouring, and the corresponding estimability graph are presented in Figure 6. Let DS3 denote
the design that coincides with the graph colouring and thus accommodates S3 . With profile set
h4, 2, 1i, DS3 gives estimates of 14 interactions. The factor grouping from the colouring is
{(B, E, F, G), (A, C), (D)}. Set S4 does not satisfy any of the conditions of Theorem 5. The
requirements graph for S4 , displayed in Figure 7, is 4-chromatic. No 27 factorial in blocks of size
four provides estimates of all main effects and all interactions in S4 . Theorem 5 can be used
to give guidance on subsets of S4 that can be accommodated. The removal of any interaction
involving at least one of C, D, G gives an interaction set that satisfies one or both of conditions
(ii) and (iii).
Sets S1 − S3 each satisfy exactly one condition of Theorem 5. It is interesting to consider
estimability properties of the two 27 factorials in blocks of size four, labelled 7-0.1/B5.1 and
7-0.1/B5.2 in Sun et al. (1997). After some work it can be shown that these designs have profile
sets h3, 2, 2i and h3, 3, 1i and provide estimates of all main effects and of 16 and 15 interactions,
respectively. Thus DS1 and DS2, of Example 4 (part 1), are isomorphic to 7-0.1/B5.1. Design,
DS3 is not isomorphic to either 7-0.1/B5.1 or 7-0.1/B5.2. Further, neither of these designs can
accommodate interaction set S3 , even with a relabelling of factors, since the requirements graph
for S3 does not admit a 3-colouring consistent with profile sets h3, 2, 2i or h3, 3, 1i.
Every acyclic graph with at least one edge is 2-chromatic and therefore 3-colourable. A result
of Bollobás and Guy (1983) establishes that many acyclic graphs are equitably 3-colourable. The
result is reproduced without proof, expressed in terms of an interaction set for an n factor process,
rather than in the graph theory terms of Bollobás and Guy:

14
 Figure 6: Graphs for DS3 of Example 5 (part 3) with interaction set S3
B1
B1 A2

C2
E1
E1
D3
A2 D3 F1
F1
G1 C2
G1
Estimability graph
Requirements graph

Figure 7: Requirements graph for interaction set S4 of Example 5 (part 3)

D
B
F

G
C
E

Theorem 6 Bollobás and Guy Let S denote a subset of interactions for an n factor process
such that there is no subset of factors Fi , for i = 1, . . . , m ≥ 3, for which all m interactions
F1 F2 , F2 F3 , . . . Fm−1 Fm , Fm F1 ∈ S. Then a design comprising a 2n factorial in blocks of size
four which provides the required estimates and achieves the bound φmax of Theorem 2 exists if
one of the following conditions is satisfied:
(i) S contains no more than (n + 8)/3 interactions involving any one factor;
(ii) The largest number of interactions involving any factor is (n + 10)/3.

Corollary 6 Consider an n factor process and interaction set S with property as specified in
Theorem 6. If S contains no more than four interactions involving any factor then a 2n factorial
in blocks of size four exists which yields the required estimates and achieves the bound of Theorem
2.

Proof: Let n0 be the largest number of interactions in S involving a common factor. Then
n0 ≤ n − 1. For n0 ≤ 4, we have n0 < (n0 + 9)/3 ≤ (n + 8)/3 and condition (i) of Theorem 6
holds.
For a situation with an acyclic requirements graph, Theorem 6 and Corollary 6 give simple
conditions guaranteeing the existence of a design which provides estimates of all effects of interest
and achieves the bound φmax .
Example 8: A 26 replicate in blocks of size four is sought to provide estimates of all interactions
in S = {AB, AC, AD, AE, EF }.
Set S satisfies Corollary 6 and thus the requirements graph (not included) has an equitable 3-
colouring. An example of such a colouring leads to the factor grouping {(A, F ), (B, C), (D, E)}.

15
Thus, a generator matrix that yields a 26 factorial in blocks of size four which provides the
required estimates and achieves φmax is:
 
1 1 1 0 0 1
X= .
0 1 1 1 1 0

4.1 Designs in blocks of size four from 2n−p factorials of resolution at

least V
We consider the construction of blocked 2n−p factorials based on 2n−p factorials of resolution at
least V. This has the advantage that both main effects and interactions will only be aliased with
negligible effects. Therefore, C1 is satisfied for all interactions and therefore any interaction of
interest that is not confounded with blocks will be estimable. For given (n, p) pair for which
2n−p factorials of resolution at least V exist, the approach is to work systematically through 2n−p
factorials of appropriate resolution in order of aberration, starting with a fractional factorial of
minimum aberration, to identify blocked 2n−p factorials covering as many profile sets as possible.
In this way we build up a portfolio of design templates. Existing catalogues are a source of 2n−p
factorials in order of aberration and details of 2n−p factorials used are contained in Table 4 in
the Appendix. The process is demonstrated by the following example.
Example 9: Design templates are sought for 27−1 factorials in blocks of size four to yield esti-
mates of all main effects and a subset of interactions.
For n = 7, the profile sets with n3 > 0 are h3, 2, 2i, h3, 3, 1i, h4, 2, 1i and h5, 1, 1i. The
27−1 factorial with minimum aberration is 7-1.1 with fraction generator F1 F2 F3 F4 F5 F6 F7 . Thus
Z = (1, 1, 1, 1, 1, 1). Potential generator matrices are constructed as follows: without loss of gen-
erality (1, 0)T is assigned to the first column of X I . All possible 35 allocations of columns of X2
to columns 2 − 6 of X I are considered. For each allocation, the 2 × 1 sub-matrix X II is obtained
as the sum of the columns of X I , working modulo 2. An allocation that yields a column of X2
as X II signifies a blocked 27−1 replicate from which all main effects are estimable. The factor
grouping is stored if the profile set has not already been identified. The scanning exercise is under-
taken using Matlab R2017a. The fractional factorial 7-1.1 yields generator matrices with profile
sets h5, 1, 1i and h3, 3, 1i. The factor groupings recorded are {(F1 , F2 , F3 , F4 , F5 ), (F6 ), (F7 )} and
{(F2 , F3 , F4 ), (F5 , F6 , F7 ), (F1 )}. To accommodate other profile sets, attention moves to the frac-
tional factorial 7-1.2, which has fraction generator F1 F2 F3 F4 F5 F7 , giving Z = (1, 1, 1, 1, 0, 1).
This yields generator matrices with X II having non-zero sum, for profile sets h4, 2, 1i and h3, 2, 2i.
The factor groupings recorded are {(F2 , F3 , F4 , F5 ), (F1 , F7 ), (F6 )} and {(F2 , F3 , F6 ), (F1 , F7 ), (F4 , F5 )}.
No further 27−1 factorials are considered since a generator matrix has now been found for every
profile set with n3 > 0. Any profile set with only two positive terms can be subsumed into one of
these sets.
It is informative to return again to Example 5 to illustrate use of the design templates.
Example 5 (part 4): Blocked 27−1 factorials, with blocks of size four, are required to provide
estimates of interactions in sets S1 , S2 and S3 of Example 5 (parts 1,3).
Denote the planned designs by DS11, DS21 and DS31. The graphs in Figures 3, 4 and 6
from Example 5 (parts 1,3) still apply. As with DS1, design DS11 can be constructed with
profile set h3, 2, 2i and factor grouping {(B, D, G), (A, E), (C, F )}. A mapping of A, . . . , G onto
F1 , . . . , F7 which is consistent with the factor grouping recorded for profile set h3, 2, 2i in Example
8 is: A → F1 , E → F7 , C → F4 , F → F5 , B → F2 , D → F3 , G → F6 , and the defining word
is ABCDEF . Design DS21 is also constructed with profile set h3, 2, 2i. With factor grouping

16
{(A, D, F ), (B, G), (C, E)}, using Example 9 a suitable mapping is: B → F1 , G → F7 , C →
F4 , E → F5 , A → F2 , D → F3 , F → F6 and the defining word is ABCDEG. Design DS31 has
profile set h4, 2, 1i and factor grouping {(B, E, F, G), (A, C), (D)}. A mapping consistent with
Example 9 is: D → F6 , A → F1 , C → F7 , B → F2 , E → F3 , F → F4 , G → F5 and the defining
word is ABCEF G.
The designs are constructed from knowledge of the factor grouping and defining word. For
example, the treatment combinations in DS31 are those in the principal 27−1 factorial with
defining word ABCEF G. A suitable generator matrix is:
 
1 0 1 1 0 0 0
X= ,
0 1 0 1 1 1 1

giving principal block (1), acd, bdef g, abcef g. The remaining 15 blocks contain the other treat-
ment combinations of the half replicate, with each block containing a coset of the treatment
combinations of the principal block.
For 5 ≤ n ≤ 11, Tables 1 and 2 give design templates for blocked 2n−p factorials with up to
128 runs using 2n−p factorials with resolution at least V. Table 1 encompasses all profile sets for
(n, p) pairs (6, 1), (7, 1) and (8, 2), and Table 2 covers all profile sets for pairs (8, 1) and (9, 2).
For pair (5, 1), Table 1 gives a design template for the profile set h3, 1, 1i. No 25−1 factorial in
blocks of size four exists with the same estimability capability as Example 1, which has profile
set h2, 2, 1i. For pair (10, 3), Table 2 gives templates for all profile sets except h8, 1, 1i, and for
pair (11, 4), templates are given for all profile sets with every factor group containing at least
two members. Use of the tables is demonstrated:
Example 10: A 28−2 factorial arranged in blocks of size four is required to assess all interactions
involving factor B and interactions AC, CH, DG, EG.
Since 28−2 factorials in blocks of size four are given for all profile sets in Table 1, the problem
is similar to that of arranging a full 28 factorial in blocks of size four. The set of interactions
satisfies conditions (ii) and (iii) of Theorem 5 and so can be represented by a requirements
graph which is 3-chromatic. We start by producing the requirements graph (see Figure 8) and
investigating possible 3-colourings. Since all interactions involving B are required, the colour
used for vertex B cannot be used for any other vertex. Thus, n3 = 1. Of the profile sets having
this property, h4, 3, 1i gives a design with the largest number of estimable interactions. There are
several colourings which result in profile set h4, 3, 1i. One such colouring is included in Figure
8 and has factor grouping {(A, D, E, H), (C, F, G), (B)}. Using Table 1, a blocked 28−2 factorial
can be constructed via mapping B → F1 , C → F2 , F → F6 , G → F8 , A → F3 , D → F4 , E →
F5 , H → F7 and defining words ABCDH and BCEF G. A generator matrix is
 
1 0 1 1 1 1 1 1
X= ,
0 1 1 0 0 1 1 0

giving principal block (1), acdef gh, bcf g, abdeh. The remaining 15 blocks are constructed as cosets
to give a design in 64 runs.

4.2 Designs in blocks of size four from 2n−p factorials of resolution IV

We give brief consideration to the construction of blocked 2n−p factorials based on fractional
factorials of resolution IV. As in §4.1, main effects will only be aliased with effects which are
negligible. However, some two factor interactions will be aliased with each other. To be specific,
each four factor interaction in the treatment defining contrast sub-group will correspond to three
pairs of aliased two factor interactions. We focus on blocked 2n−p factorials obtained from 2n−p

17
 Figure 8: Requirements graph for Example 10
B3 C2

A1 D1

H1 E1

G2 F2

Table 1: Templates for 2n−p factorials in blocks of size four with up to

64 runs, from 2n−p factorials of resolution at least V
n p runs profile inta fractional factor
set factorial grouping
5 1 16 h3, 1, 1i 7 5-1.1 {(F1 , F2 , F3 ), (F4 ), (F5 )}
6 1 32 h2, 2, 2i 12 6-1.1 {(F1 , F6 ), (F2 , F3 ), (F4 , F5 )}
6 1 32 h3, 2, 1i 11 6-1.2 {(F2 , F3 , F6 ), (F4 , F5 ), (F1 )}
6 1 32 h4, 1, 1i 9 6-1.2 {(F1 , F2 , F3 , F5 ), (F4 ), (F6 )}
7 1 64 h3, 2, 2i 16 7-1.2 {(F2 , F3 , F6 ), (F1 , F7 ), (F4 , F5 )}
7 1 64 h3, 3, 1i 15 7-1.1 {(F2 , F3 , F4 ), (F5 , F6 , F7 ), (F1 )}
7 1 64 h4, 2, 1i 14 7-1.2 {(F2 , F3 , F4 , F5 ), (F1 , F7 ), (F6 )}
7 1 64 h5, 1, 1i 11 7-1.1 {(F1 , F2 , F3 , F4 , F5 ), (F6 ), (F7 )}
8 2 64 h3, 3, 2i 21 8-2.1 {(F1 , F4 , F7 ), (F2 , F5 , F6 ), (F3 , F8 )}
8 2 64 h4, 2, 2i 20 8-2.1 {(F1 , F2 , F3 , F5 ), (F4 , F6 ), (F7 , F8 )}
8 2 64 h4, 3, 1i 19 8-2.1 {(F3 , F4 , F5 , F7 ), (F2 , F6 , F8 ), (F1 )}
8 2 64 h5, 2, 1i 17 8-2.1 {(F2 , F4 , F5 , F6 , F7 ), (F3 , F8 ), (F1 )}
8 2 64 h6, 1, 1i 13 8-2.1 {(F3 , F4 , F5 , F6 , F7 , F8 ), (F1 ), (F2 )}
a
number of estimable interactions

18
Table 2: Templates for 2n−p factorials in blocks of size four with 128 runs, from
2n−p factorials of resolution at least V
n p profile inta fractional factor
set factorial grouping
8 1 h3, 3, 2i 21 8-1.2 {(F2 , F3 , F4 ), (F5 , F6 , F8 ), (F1 , F7 )}
8 1 h4, 2, 2i 20 8-1.1 {(F2 , F3 , F4 , F5 ), (F1 , F8 ), (F6 , F7 )}
8 1 h4, 3, 1i 19 8-1.2 {(F2 , F3 , F4 , F7 ), (F5 , F6 , F8 ), (F1 )}
8 1 h5, 2, 1i 17 8-1.2 {(F2 , F3 , F4 , F5 , F8 ), (F6 , F7 ), (F1 )}
8 1 h6, 1, 1i 13 8-1.2 {(F1 , F2 , F3 , F4 , F5 , F7 ), (F6 ), (F8 )}
9 2 h3, 3, 3i 27 9-2.1 {(F1 , F8 , F9 ), (F2 , F4 , F6 ), (F3 , F5 , F7 )}
9 2 h4, 3, 2i 26 9-2.1 {(F2 , F3 , F6 , F9 ), (F1 , F7 , F9 ), (F4 , F5 )}
9 2 h5, 2, 2i 24 9-2.1 {(F1 , F6 , F7 , F8 , F9 ), (F2 , F3 ), (F4 , F5 )}
9 2 h4, 4, 1i 24 9-2.2 {(F1 , F4 , F8 , F9 ), (F2 , F5 , F6 , F7 ), (F3 )}
9 2 h5, 3, 1i 23 9-2.2 {(F3 , F5 , F6 , F7 , F9 ), (F1 , F2 , F8 ), (F4 )}
9 2 h6, 2, 1i 20 9-2.2 {(F2 , F4 , F5 , F6 , F7 , F8 ), (F1 , F9 ), (F3 )}
9 2 h7, 1, 1i 15 9-2.2 {(F1 , F2 , F3 , F5 , F6 , F7 , F9 ), (F4 ), (F8 )}
10 3 h4, 3, 3i 33 10-3.1 {(F3 , F5 , F7 , F10 ), (F1 , F8 , F9 ), (F2 , F4 , F6 )}
10 3 h4, 4, 2i 32 10-3.1 {(F1 , F6 , F8 , F10 ), (F2 , F3 , F7 , F9 ), (F4 , F5 )}
10 3 h5, 3, 2i 31 10-3.1 {(F2 , F3 , F6 , F9 , F10 ), (F1 , F7 , F8 ), (F4 , F5 )}
10 3 h5, 4, 1i 29 10-3.1 {(F2 , F4 , F6 , F7 , F9 ), (F1 , F3 , F5 , F8 ), (F10 )}
10 3 h6, 2, 2i 28 10-3.1 {(F1 , F6 , F7 , F8 , F9 , F10 ), (F2 , F3 ), (F4 , F5 )}
10 3 h6, 3, 1i 27 10-3.1 {(F2 , F3 , F4 , F6 , F7 , F9 ), (F1 , F5 , F9 ), (F10 )}
10 3 h7, 2, 1i 23 10-3.1 {(F2 , F4 , F5 , F6 , F7 , F8 , F9 ), (F1 , F3 ), (F10 )}
11 4 h4, 4, 3i 40 11-4.1 {(F2 , F4 , F6 , F11 ), (F3 , F5 , F7 , F10 ), (F1 , F8 , F9 )}
11 4 h5, 3, 3i 39 11-4.1 {(F2 , F4 , F6 , F7 , F9 ), (F1 , F3 , F11 ), (F5 , F8 , F10 )}
11 4 h5, 4, 2i 38 11-4.1 {(F2 , F3 , F6 , F9 , F10 ), (F1 , F7 , F8 , F11 ), (F4 , F5 )}
11 4 h6, 3, 2i 36 11-4.1 {(F4 , F5 , F6 , F7 , F10 , F11 ), (F1 , F8 , F9 ), (F2 , F3 )}
11 4 h7, 2, 2i 32 11-4.1 {(F1 , F6 , F7 , F8 , F9 , F10 , F11 ), (F2 , F3 ), (F4 , F5 )}
a
number of estimable interactions

19
factorials of resolution IV for which the treatment defining contrast subgroup contains only one
word of length four. Since the columns of X pertaining to this four factor interaction necessarily
sum to (0, 0)T , the four columns are either four copies of the same column of X2 or two columns
of X2 each copied twice. Thus, the four constituent factors of the interaction will all be contained
in one factor group or will be distributed between two factor groups with each containing a pair.
In the former situation, the factors of the aliased two factor interactions occur in the same factor
group. Thus, the interactions fail C2 and so also failing C1 has no impact on the estimability
capability of the design. In the latter case, four interactions involving factors in two factor groups
will be inestimable: these will fail only C1.
Table 3 gives design templates for n = 7, 9, 12 not covered in Tables 1 and 2. Where a design
can be found for an (n, p) pair and profile set combination with all aliased two-factor interactions
involving factors in the same factor group then such a design is the only one recorded, since it
improves on designs with aliased interactions spread between two factor groups. See for example
the blocked 27−2 factorial with profile set h1, 1, 5i in the third row of Table 3. Designs with
common (n, p) pair and profile set combination with aliased interactions spread between two
factor groups do not necessarily have equivalent estimability structure. For example, consider
the two blocked 29−3 factorials both with profile set h4, 3, 2i in the seventh and eighth rows
of Table 3. Both have the factors of aliased interactions split between two factor groups but
the estimability graphs are not isomorphic. Both estimability graphs are formed from the same
complete 3-partite graph, with four edges deleted. In the first case the deleted edges are between
the vertex sets (factor groups) of sizes two and four, whereas in the second they are between the
vertex sets (factor groups) of sizes three and four.
Designs in which factors involved in aliased two-factor interactions are all contained in the
same factor group are used in exactly the same way as those in Tables 1 and 2. For an (n, p) pair
and profile set with aliased two-factor interactions involving two factor groups, care has to be
given to allocating factors within factor groups to take account of the non-estimable interactions,
where possible. To illustrate the consequence of alternative graph colourings, and of different
mappings within factor groups for a given colouring, we return one final time to Example 5.
Example 5 (part 5): A design comprising a 27−2 factorial in blocks of size four is sought to
provide estimates of interactions in set S2 .
The requirements graph and colouring in Figure 4 correspond to profile set h3, 2, 2i and factor
grouping {(A, D, F ), (B, G), (C, E)}. The top row of Table 3 suggests a mapping: B → F1 , G →
F6 , C → F2 , E → F3 , A → F4 , D → F5 , F → F7 with defining words BCEG and ABCDF .
However, four of the selected interactions, namely BC, BE, CG and EG, are aliased with other
two-factor interactions and so are not estimable. Bar a reordering of the colour labels, Figure
4 gives the only colouring with profile h3, 2, 2i. Thus, the 32 run design will provide estimates
of all main effects but of only six of the interactions in S2 . We consider colourings leading to
alternative profile sets. Figure 5 gives a colouring for the requirements graph of S2 consistent with
profile set h3, 3, 1i, and factor grouping {(A, D, E), (B, F, G), (C)}. The construction from the
second row of Table 3 will give a suitable design but with interactions involving factors mapped
onto F1 − F4 aliased. To construct such a design, two factors from the factor group comprising
A, D, E are mapped onto F1 , F2 and two factors from B, F, G are mapped onto onto F3 , F4 . The
estimability graphs for three mappings are illustrated in Figure 9. Each consists of a complete
3-partite graph consistent with profile set h3, 3, 1i with four edges removed. The edges to be
removed relate to interactions that are inestimable despite involving factors from different factor
groups and are indicated by dotted lines for clarity. The mappings in Figure 9 (i) and (ii) lead
to two and three interactions of S2 being inestimable. The final mapping of Figure 9 (iii) is
preferred over these since it provides all the required estimates. The full mapping of Figure 9 (iii)

20
 Figure 9: Estimability graphs for blocked 27−2 factorials with interaction set S2
A1 B2 A1 B2 A1 B2

C3 C3 C3

D1 G2 D1 G2 D1 G2

E1 F2 E1 F2 E1 F2

(i) A → F1 , D → F2 (ii) A → F1 , E → F2 (iii) A → F1 , D → F2

B → F3 , G → F4 B → F3 , G → F4 F → F3 , G → F4
AB, BD aliased AB, BE, EG aliased no S2 terms aliased

is C → F5 , A → F1 , D → F2 , E → F7 , B → F6 , F → F3 , G → F4 . A 27−2 factorial in blocks of

size four consistent with this has generator matrix:
 
1 0 1 1 1 0 0
X= ,
0 1 1 0 0 1 1

and defining words ABEF and ACDEG. This shows that a profile set able to give estimates of
fewer interactions can be advantageous and that different mappings result in different numbers
of interactions of S2 being estimable.

5 Concluding Remarks
This work develops a practitioner led approach to the construction of blocked 2n and 2n−p
designs to estimate main effects and selected two-factor interactions. The generator matrix
provides immediate and illuminating information on the capacity of a design to estimate the
chosen effects. Hence, the construction focus on the generator matrix, rather than on the block
defining contrast sub-group, facilitates the building of bespoke designs. The method is especially
advantageous and convenient when the block size is relatively small. It is notable that the most
useful blocked 2n factorials are not necessarily those which maximise the number of estimable
interactions. Likewise, the most useful blocked 2n−p factorials are not necessarily based on
minimum aberration 2n−p fractions.
A novel aspect of the work is the use of concepts from graph theory. The representation of a
blocked 2n factorial by a requirements graph with proper vertex colouring enables exploitation
of results on chromatic numbers and proper vertex colourings.
The design class with blocks of size four is an important one and is the class which gives
the most construction challenges. The difficulty is, in part, due to the large number of factorial
effects that are confounded with blocks. The other feature of blocks of size four that contributes
to the design limitations is the existence of small configurations of interactions that cannot be
accommodated: any interaction set containing all six pairwise interactions between four factors
will correspond to a requirements graph with chromatic number at least four. No design can be
found to cover all main effects and such an interaction set. For fractional designs The tables of
designs provided for designs in blocks of size four should form a useful resource.

21
Table 3: Templates for blocked 2n−p factorials with up to 128 runs, from 2n−p factorials of resolution IV
n p runs profile inta fractional factor aliasedb
set factorial grouping interactions
7 2 32 h3, 2, 2i 12 7-2.1 {(F4 , F5 , F7 ), (F1 , F6 ), (F2 , F3 )} F1 F2 , F1 F3 , F2 F6 , F3 F6
7 2 32 h3, 3, 1i 11 7-2.1 {(F1 , F2 , F7 ), (F3 , F4 , F6 ), (F5 )} F1 F3 , F1 F6 , F2 F3 , F2 F6
7 2 32 h5, 1, 1i 11 7-2.1 {(F1 , F2 , F3 , F4 , , F7 ), (F5 ), (F7 )} none
7 2 32 h4, 2, 1i 10 7-2.1 {(F1 , F5 , F6 , F7 ), (F2 , F3 ), (F4 )} F1 F2 , F1 F3 , F2 F6 , F3 F6
9 3 64 h5, 2, 2i 24 9-3.1 {(F1 , F2 , F3 , F4 , F7 ), (F5 , F6 ), (F8 , F9 )} none
9 3 64 h3, 3, 3i 23 9-3.1 {(F1 , F3 , F9 ), (F2 , F6 , F7 ), (F4 , F5 , F8 )} F1 F2 , F1 F7 , F2 F3 , F3 F7
9 3 64 h4, 3, 2i 22 9-3.1 {(F1 , F5 , F7 , F8 ), (F4 , F6 , F9 ), (F2 , F3 )} F1 F2 , F1 F3 , F2 F7 , F3 F7
9 3 64 h4, 3, 2i 22 9-3.1 {(F2 , F4 , F5 , F7 ), (F1 , F3 , F9 ), (F6 , F8 )} F1 F2 , F1 F7 , F2 F3 , F3 F7
9 3 64 h6, 2, 1i 20 9-3.1 {(F1 , F2 , F3 , F7 , F8 , F9 ), (F5 , F6 ), (F4 )} none
12 5 128 h4, 4, 4i 48 12-5.1 {(F1 , F8 , F9 , F12 ), (F2 , F4 , F6 , F11 ), none
(F3 , F5 , F7 , F10 )}
12 5 128 h5, 4, 3i 47 12-5.2 {(F3 , F4 , F6 , F10 , F12 ), (F2 , F5 , F7 , F11 ), none
(F1 , F8 , F9 )}
12 5 128 h6, 4, 2i 44 12-5.1 {(F4 , F5 , F6 , F7 , F10 , F11 ), (F1 , F8 , F9 , F12 ), none
(F2 , F3 )}
12 5 128 h6, 3, 3i 41 12-5.2 {(F4 , F5 , F6 , F7 , F10 , F11 ), (F1 , F8 , F9 ), F3 F4 , F3 F6 , F4 F12 , F6 F12
(F2 , F3 , F12 )}
12 5 128 h6, 3, 3i 41 12-5.2 {(F1 , F2 , F5 , F8 , F10 , F11 ), (F3 , F4 , F9 ), F3 F6 , F3 F12 , F4 F6 , F4 F12
(F6 , F7 , F12 )}
12 5 128 h5, 5, 2i 41 12-5.2 {(F1 , F3 , F5 , F8 , F12 ), (F2 , F4 , F6 , F7 , F9 ), F3 F4 , F3 F6 , F4 F12 , F6 F12
(F10 , F11 )}
12 5 128 h7, 3, 2i 37 12-5.3 {(F2 , F3 , F4 , F6 , F9 , F11 , F12 ), (F1 , F7 , F8 ), F1 F2 , F1 F3 , F2 F8 , F3 F8
(F5 , F10 )}
12 5 128 h8, 2, 2i 36 12-5.1 {(F1 , F6 , F7 , F8 , F9 , F10 , F11 , F12 ), (F2 , F3 ), none
(F4 , F5 )}
a
number of estimable interactions
b
interactions with constituent factors from different factor groups which are aliased with another inter-
action

22
 Table 4: 2n−p Factorials
fractional defining words
factorial
5-1.1 F1 F2 F3 F4 F5
6-1.1 F1 F2 F3 F4 F5 F6
6-1.2 F1 F2 F3 F4 F6
7-1.1 F1 F2 F3 F4 F5 F6 F7
7-1.2 F1 F2 F3 F4 F5 F7
7-2.1 F1 F2 F3 F6 , F1 F2 F4 F5 F7 ,
8-1.1 F1 F2 F3 F4 F5 F6 F7 F8
8-1.2 F1 F2 F3 F4 F5 F6 F8
8-2.1 F1 F2 F3 F4 F7 , F1 F2 F5 F6 F8
9-2.1 F1 F2 F3 F4 F5 F8 , F1 F2 F3 F6 F7 F9
9-2.2 F1 F2 F3 F4 F8 , F1 F2 F5 F6 F7 F9
9-3.1 F1 F2 F3 F7 , F1 F2 F4 F5 F8 , F1 F3 F4 F6 F9
10-3.1 F1 F2 F3 F4 F5 F8 , F1 F2 F3 F6 F7 F9 , F1 F2 F4 F6 F10
11-4.1 F1 F2 F3 F4 F5 F8 , F1 F2 F3 F6 F7 F9 , F1 F2 F4 F6 F10 , F1 F3 F5 F7 F11
12-5.1 F1 F2 F3 F4 F5 F8 , F1 F2 F3 F6 F7 F9 , F1 F2 F4 F6 F10 , F1 F3 F5 F7 F11 , F1 F4 F5 F6 F7 F12
12-5.2 F1 F2 F3 F4 F5 F8 , F1 F2 F3 F6 F7 F9 , F1 F2 F4 F6 F10 , F1 F3 F5 F7 F11 , F3 F4 F6 F12
12-5.3 F1 F2 F3 F8 , F1 F2 F4 F5 F9 , F1 F3 F4 F6 F10 , F2 F3 F5 F7 F11 , F4 F5 F6 F7 F12

A Appendix
The 2n−p factorials listed in Table 4 are obtained from Chen et al. (1993) and Block and Mee
(2005).

References
Block, R.M. and Mee, R.W. (2005) Resolution IV designs with 128 runs. Journal of Quality
Technology 37, 282–293.
Bollobás, B., Guy, R.K. (1983) Equitable and proportional coloring of trees. Journal of Combi-
natorial Theory, Series B 34, 177–186.
Bondy, J.A. and Murty, U.S.R. (2008) Graph Theory. Springer.
Brooks, R. L. (1941) On colouring the nodes of a network. Proc. Cambridge Phil. Society 37
194–197.
Chen, H. and Cheng, C.-S. (1999). Theory of optimal blocking of 2n−m designs. Annals of
statistics 27, 1948–1973.
Chen, J., Sun, D. X. and Wu, C. F. J. (1993) A catalogue of two-level and three-level fractional
factorial designs with small runs. International Statistical Review 61, 131–145.

Cheng, C.-S. and Mukerjee, R. (2001). Blocked regular fractional factorial designs with maximum
estimation capacity. Annals of statistics 29, 530–548.
Fries, A. and Hunter, W.G. (1980). Minimum aberration 2k−p designs. Technometrics 22,
601–608.

23
 Godolphin, J.D. (2019). Construction of row-column factorial designs. Journal of the Royal
Statistical Society, Series B 81, 335–360.
Hedayat, A.S. and Pesotan, H. (1992) Two-level factorial designs for main effects and selected
two-factor interactions. Statistica Sinica 2, 453–464.
Mead, R., Gilmour, S.G. and Mead, A. (2012) Statistical principles for the design of experiments.
Applications to real experiments. Cambridge University press.
Mukerjee, R. and Wu, C.F.J. (1999). Blocking in fractional factorials: a projective geometry
approach. Annals of statistics 27, 1256–1271.
Sitter, R.R., Chen, J. and Feder, M. (1997) Fractional resolution and minimum aberration in
blocked 2n−k designs. Technometrics 39 382–390.
Stein, W. et al. (2018) Sage Mathematics Software (Version 8.2). The Sage Development Team,
http://www.sagemath.org.
Sun, D. X., Wu, C. F. J. and Chen, Y. (1997) Optimal blocking schemes for 2n and 2n−p designs.
Technometrics 39, 298–307.
Wang, P.C. (2007). Planning experiments when some specified interactions are investigated.
Computational Statistics and Data Analysis 51 4143 – 4151.
Wu, C.F.J. and Chen, C. (1992) A graph-aided method for planning two-level experiments when
certain interactions are important. Technometrics 34 162–175.
Wu, C.F.J. and Hamada, M.S. (2009) Experiments: planning, analysis, and optimization (second
edition). Wiley, New York.
Wu, H., Mee, R. and Tang, B. (2012) Fractional factorial designs with admissible sets of clear
two-factor interactions. Technometrics 54 191–197.
Xu, H. (2006) Blocked regular fractional factorial designs with minimum aberration. Annals of
Statistics, 34, 2534–2553.
Xu, H. and Lau, S. (2005) Minimum aberration blocking schemes for two- and three-level
fractional factorial designs. Journal of Statistical Planning and Inference, 136, 4088–4118.
Xu, H. and Mee, R.W. (2010) Minimum aberration blocking schemes for 128-run designs. Journal
of Statistical Planning and Inference, 140, 3213–3229.
Zhang, R. and Park, D. (2000) Optimal blocking of two-level fractional factorial designs. Journal
of Statistical Planning and Inference, 91, 107–121.

24

View publication stats