The impact of building design on maintenance costs of residential housing estates owned by Mombasa municipal council
Abstract
This study constitutes an investigation of the
impact of building design on the subsequent costs of
maintaining buildings in use. It s restricted to
residential buildings owned and maintained by Mombasa
Municipal Council which is one of the oldest urban
authorities in Kenya, and is the second largest urban
authority after Nairobi. The study is solely concerned
with the effect of building morphology (size), construction
materials, and density on maintenance costs. It is assumed
that these factors are within the control of the designer,
so that their manipulation can either increase or decrease
the total amount of money used in both the construction
and maintenance of a building t hroughout its total life.
Thus the designer's role in this study is seen as that of
sole adviser to building owners as to what would be appropriate
economically as well as to users.
The study, however recognises that during the use,
of a building other factors such as management policy,
finance, building age, mode of use, climate etc. may in
various degrees affect not only the maintenance needs of
a building but also how well the needs can be satisfied.
Management policy and climate are,assumed constant in this .
study since all the houses belong to one organisation,
and within one climatic region. Other factors may vary
within different hciusing estates, and therefore may be
measurable.
Given the complexity of the factors that may
influence maintenance costs, it has been found necessary
to employ the technique of multiple linear regression
analysis to measure the degree of association between any
of the factors and maintenance costs. Therefore the
study uses a simple theoretical mathematical model of the
f6rm Y = f(XI, X2 Xn) as framework within which to
study how the annual maintenance costs vary with building
design and other relevant factors. For the purposes of
this model, 19 independent variables were identified for
regression against maintenance cost as the dependent.viiiable.
The study has been organised into seven chapters.
The first chapter introduces. the prpblems identified in the
study area, the study objectives, as well as the scope and
definition of important terms used in the study. It is
noted in this chapter that housing estates in Kenya are
not only deteriorating fast, but they are also expensive to
maintain. Questions were raised, for instance, why some
local authorities should spend as much as 58.3% of their
rental income on maintenance of their housing estates, when
infact, only l5%is what is allowed for during estate
development appraisal. Whether the high costs arise from
poor management or bad design, the consequences are
disastrous to the whole country whose capacity to build new
houses as rapidly as existing ones become unusable, is
insufficient.
Chapter Two attempts to'set a theoretical framework
from some known or assumea principles within which to
investigate factors affecting maintenance costs of
residential housing estates. The framework leads to the
formulation of a null hypothesis that no relationship
exists between building design and maintenance costs. A conceptual maintenance cost model in the form of multiple
linear regression is developed to test the validity or
otherwise of the above hypothesis. The chapter ends with
a discussion on how the research was carried but in
the field and how the data would be analysed and presented.
The third chapter discusses representative literature
selected from various references on the economics of
building maintenance, and the influence of design on
I maintenance costs. It was found in reviewing the literature
for this study that thereis no cost model that includes
maintenance expenditure as a design criterion; It was also
found that there is no agreed method of determining how
much money should be set aside for maintenance costs.
The chapter, concludes that there is need to cohsider.
the significance of maintenance costs during building
design. The chapter recommends that in order to determine
maintenance needs of a building, it is necessary to use
the concept of duty of care which relates the building to
a hospital patient.
the application of this concept is
further discussed in
process is described.
Chapters Four, Five and Six deal with the case study
area of Mombasa Municipal Council. Chapter Four consists
of general background to the study area'of Mombasa and
discusses the factors determining house design in this
hot and humid climatic zone. The chapter finds that
Mombasa is essentially an African town since Africans
comprise 76% of the population. It also notes that one
single important factor in house design is that it should
satisfy people's way of life. It does appear, however,
that the traditional "Swahili-house" is more amenab'le to
satisfying the needs of the majority of Mombasa's population
in comparison to corrveniional house provided by
Mombasa Municipal Council. This house is generally large
in accordance with the generally' large African households. I
and it also lends itself to easy construction techniques. .
It satisfies the climatic criterion of cross ventilation.
It is therefore concluded that its features should be
adopted to the extent possible in the conventiorial council
house.
Chapter Five examines the incidence of maihtenahce
failure characteristics. A number of quantitative
aspects of maintenance management process such as frequency:
or volume of work orders were investigated, as well as
qualitative aspects such as audit conditions and the use
of photography to give explanations to defects that create
maintenance. It is concluded in this chapter-that houses
in Mombasa are poorly maintained, and that there is a
backlog in the order of 42% of reported defects uhattended.
While maintenance management may be the major cause of
this backlog, it has been recognised that initial design
created fertile grounds for the need for frequency of
maintenance. The sixth chapter evaluates the maintenance
costs collected in terms of building design features and
other relevant factors. It is observed in this chapter
that wall decoration and plumbing works are the most
expensi ve .el.emen t s . in Mombasa. This is largely
because of pain twork on external' surfaces on buildings
which calls for frequent redecoration,and the use of
steel pipes externally. When the maintenance cost data
were subjected to statistical analysis, about seven
independent variables out of 19 variables were found
significant at a 95% confidence limit. Rental value obtained
for each house was the most signficantsingle factor
accounting, for more than 86% of the variation in maintenance
cost per 'accommodation unit and irnore than 68% of the'
maintenance cost per room. The estates with high rents
cost more to maintain in 'comparsion to low rent estates
thus confirming that finance is important in building
maintenance. Other signiftcant variables included perimeter
length, room height, number of housing units per
estate, number of habitable rooms per hectare, .type of
roof and type of wall construction. The variables in all
accounted for 99.3% of the variation in maintenance costs
per unit of accommodation. The plot of standardised
restduals was within less than one standard deviation,
showing that the model nearly predicts the observations.
The last chapter is a summary of the findings of
the study and recommendations for further research into
the problems of building maintenance. It is concluded
that a relationship exists between building design and
maintenance costs in terms of choice of materials building
density and building shape. Recommendations are made
regarding appropriate use of building materials density
and need for compact buildings. Further recommendations
include need to externalise design and the need for
improved methods for maintenance management. It is
however, recognised -that the model developed for Mombasa
Council housing estates may not be universally applicable
throughout the country, but that the variables so
identified should be tested in other areas to verify their
behaviour with respect to maintenance costs. Climatic,
variations as well as management policies of various houslhg
authorities may need further investigations before
general conclusions can be drawn on the operation of
maintenance cost models for residential buildings.
Citation
Degree of Doctor of PhilosophyPublisher
University of Nairobi Department of Land Development
Description
A thesis submitted in partial fulfilment
for the Degree of Doctor of
Philosophy in the Department
of Land Development at the
University of Nairobi.