MODELLING THE VALUE OF CONSTRUCTION MECHARONICS INVESTMENT

OPPORTUNITIES

M D Taylor
School of the Built Environment, Napier University, Edinburgh, EH10 5DT
E-mail: [email protected]

ABSTRACT: Deficiencies in the existing procedures for assessing the financial feasibility
of construction automation have manifested themselves that further research is required to
reveal the true economic potential of available construction mechatronics systems. The
author's PhD research highlights the deficiencies in existing appraisal techniques and
develops a more befitting valuation framework. Probabilistic risk analyses are
implemented to describe uncertain project cash flows associated with existing and
prototype systems. The research examines the possible choice of input probability
distribution, the interdependence between input variables and the sensitivity of the decision
criterion to the input variables. Furthermore, real option-pricing models are developed to
encapsulate the managerial options available to plant-hire organisations and contractors.

Keywords: construction mechatronics, strategic valuation

1. INTRODUCTION

The essence of probabilistic risk analysis is to provide the investment analyst with a means to
look ahead at possible future outcomes and evaluate whether the investment should be approved.
Projected cash flows are often estimates of quantities whose true values are uncertain because
they will be determined in the future. The current-status of construction automation
implementation does not facilitate the estimation of uncertain costs and revenues. Vagueness
concerning cash flow estimation originates from uncertain ownership and operating expenditures.
Non-tangible benefits substantially contribute to the economic value of construction automation.
Practical experience with advanced construction automation technology is limited and further
implementation is required to generate an appropriate historical database of associated
expenditures.

Within the construction sector, risk is largely perceived as financial (Baker, Ponniah & Smith
1999). The risk associated with the introduction of new technology is also regarded as an
important area for future research (Edwards & Bowen 1998). Furthermore, as identified by the
CRISP Commission, risk and evidence of success were priority issues in reducing the barriers to
technological change (CRISP 2000). Exploration of the financial risk associated with automated
construction technology is necessary to assist quantification and understanding of investment and
utilisation risk. Recent debates have discussed whether investment decision regarding advanced
manufacturing technology can be based upon quantitative financial techniques or should be
undertaken on strategic grounds. Analysis of financial risk in conjunction with the mapping of
construction automation investments as real options may assist in assessing the existing and
future strategic value of the technology. An overview of the pertinent research issues concerning
the modelling and simulation aspects of the research are introduced. Preliminary results and a
selection of conclusions are presented.

2. OBJECTIVES & METHODOLOGY

The research aims to provide a detailed risk assessment and analysis of the possibility of plant
hire organisations investing in automated construction technology plant hire for introduction into
UK construction and civil engineering operations. The research highlights the deficiencies in

3
existing valuation methodologies and, subsequently, applies real option-pricing theory to assess
the value of strategic construction mechatronics investment opportunities.

A series of available systems are utilised as potential investments. Objective data is obtained
from the machine manufacturers or designers, where possible, and supplemented with subjective
estimates. Subjective cash flow data and real option parameters are modelled and simulated to
produce NPV and option value profiles for the selected systems. The valuation techniques and
their interaction are outlined in Figure 1. Switching, growth (compound), pioneer and timing
options are modelled and recommendations for appropriate implementation strategies are
formulated.

DECISION

RISK REAL OPTION

ANALYSIS ANALYSIS

REAL OPTION
NPV VALUE REAL OPTION MODELS
PROFILES VALUE PROFILES

INPUT
PARAMETERS

SENSITIVITY NET PRESENT

ANALYSIS VALUE MODEL

HOURLY HIRE RATE

(McCaffer 1995)

Figure 1 Construction mechatronics investment decision methodology.

3. INVESTMENT SCENARIO

Contractors aim to minimise their fixed capital locked into production by hiring or leasing
expensive plant and equipment. The construction industry avoids working practices that demand
substantial investment in plant and machinery. Construction contractors sub-contract a
substantial proportion of their work and subsequently reduce their commitment to fixed capital.
Construction projects are unique and specialist plant with operators is generally only required for
limited period. The plant hire industry meets the requirements of contractors and facilitates
access to plant and operators without exposure to maintenance and technical difficulties. With
regards to civil engineering operations, the work often involves the use of expensive plant, which,
to be cost effective, has to be utilised to its utmost. Within the research, it is assumed that a plant
hire organisation will purchase construction mechatronics (CM) technology to be hired to
construction and civil engineering contractors throughout the UK construction industry.

4. APPROPRIATE PROBABILITY DISTRIBUTIONS

The conducted financial risk analyses requires the estimation of cash flows and the construction
of subjective probability distributions. The reviewed literature outlines the difficulties associated
with utilising subjective estimations within probabilistic risk analysis.

Objective probabilities are difficult to obtain from the construction industry, where each project is
unique. If objective data is not available for determining probability distributions for inputs to a
risk analysis model, then subjective data, based upon managerial judgement, may be utilised.

4
Illogical results follow from the assumption that the input subsystems have a triangular
probability density function (Chau 1995). The triangular p.d.f systematically overestimates the
probability of exceeding the most likely estimate. The beta distribution, widely used to model
construction duration, may not be appropriate for modelling subjective data. Raftery (1994)
suggests that risk analysis is not within the realm of repeatable statistical assessment, but of
subjective definitions of probability. For reasons of simplicity, Raftery (1994) then recommends
that the uniform, triangular, trapezoidal, step-rectangular and discrete distribution for practical
use in risk analysis for construction project’s. Chau (1995) produced empirical data, which
showed that the log-triangular p.d.f provided a better approximation of real construction cost data.
Furthermore, neglecting subsystem correlation is erroneous analysis and produces misleading
results (Wall 1997).

Probability distributions will also be utilised to describe the real option-pricing valuation
parameters. The research investigates the use of the probability distribution functions suggested
within the literature. However, without objective data obtained from system users, the
appropriateness of the chosen distributions may not be validated.

5. REAL OPTION PRICING

Real option-pricing analysis recognises the additional value that arises from managerial
flexibility. Furthermore, it captures management’s flexibility to adapt and revise later decisions
in response to unexpected market developments. The managerial options available to prospective
investors may assist in highlighting the value of existing mechatronics. The Black and Scholes
formula (Black & Scholes 1973) can be used to value a European call option. A CM investment
timing option may be equated to a European call option. The value of the call option may be
calculated by mapping the proposed investment onto the Black & Scholes option parameters.

5.1 Volatility

Within the real option-pricing models, the volatility of the underlying provides an additional
source of value, which is not accounted for within traditional NPV valuations. Volatility may be
estimated from historical data (FTSE Construction & Building Materials Sector) and then a
probability distribution may be constructed to describe the uncertainty surrounding the estimate.
Alternatively, once the NPV profiles have been constructed for the financial risk analyses, the
standard deviation of the returns (i.e. the underlying) can be used as an estimation of volatility.
The different techniques available to describe volatility will be examined within the research.

5.2 The Underlying (NPV of Project Cash Flows)

Within the risk analyses the effect of changing the type of input p.d.f, the number of iterations,
altering the coefficient of variation, the estimated mean values; interdependence and the type of
simulation are examined. The results from adjusting the coefficient of variation for the input
subsystem variables within the risk analysis model are presented as cumulative ascending
distribution in Figure.2.

5
 Plant Hire Model: Effect of Varying the Input Variables
Coefficient of Variation (50000, LHS)

0.9

0.8

Probability of Value <= X-axis Value

0.7

0.6 V=2%, 50000, LHS

0.5 V=5%, 50000, LHS

0.4 V=7.5%, 50000, LHS

0.3 V=10%, 50000, LHS

0.2

0.1

0
-700

-400

-100

200

500

800
-1000

1100

1400

1700

2000

2300

2600

2900

3200

3500

3800

4100

4400

5000
NPV (£)

Figure 2 Effect of varying the coefficient of variation for all subsystem input variables.

6. CONCLUSIONS

The main conclusions derived from this study are presented as follows:

1. Probabilistic risk analysis adequately models the financial risk characteristics of CM

investment opportunities
2. Varying the coefficient of variation of the subsystem input variables affected the NPV
profile for the described investment scenario model.
3. To test whether interdependence exists and the reliability of subjective cash flow data,
objective (historical) cash flow data is required.
4. Risk analysis, real option pricing and probabilistic risk analysis may be combined to provide
a more accurate description of the value of CM investment opportunities.

7. REFERENCES

BAKER, S., PONNIAH, D. & SMITH, S. (1999) Survey of Risk Management in Major UK
Companies, Journal of Professional Issues in Engineering Education and Practice, Vol.125
(3), 94-102.
CHAU, K.W. (1995) Monte Carlo simulation of construction costs using subjective data,
Construction Management and Economics, Vol.13, 369-383.
CRISP (2000) Report on a Workshop: Technological Change and Rethinking Construction,
Construction Research and Innovation Strategy Panel, July, [http://www.crisp-uk.org.uk/].
EDWARDS, P.J. & BOWEN, P.A. (1998) Risk and risk management in construction: a review
and future directions for research, Engineering Construction and Architectural Management,
Vol.5 (4), 339-349.
RAFTERY, J. (1994) Risk Analysis in Project Management, E & FN Spon, London.
WALL, D.M. (1997) Distributions and correlation’s in Monte Carlo Simulation, Construction
Management and Economics, Vol.15, 241-258.