A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Forecasting UK construction plant sales
To sustain competitiveness, construction plant manufacturers must be able to accurately forecast market sales and moreover, their anticipated percentage of those sales. This paper addresses the former of these forecasts through the development of a multivariate time series model. Specifically, an autoregressive moving average (ARMA) time series model (otherwise known as the Box-Jenkins approach) is constructed using economic data relating to a 15-year period (1985–99). It is identified that population (millions); housing completions total (millions), total building repair and maintenance (billions Euro at 1998 prices, where 1 Euro = £0.6776) and gross domestic product (millions at market prices) are able to accurately predict plant sales. The performance statistics show the derived time series model to be very accurate in modelling the dependent variable [mean absolute deviation (MAD) = −87.04 and root mean square error (RMSE) = 1404.05]. By observing economic interactions and influences on sales, it is envisaged that the manufacturer can adjust their production output to match demand. Based upon the model produced, a forecast of machine sales for year 2000 is made and indicates that sales numbers look set to reduce to 15 482 units. The paper concludes with direction for future research work that will aim to model machine park, the influence of grey machines and individual plant items.
Forecasting UK construction plant sales
To sustain competitiveness, construction plant manufacturers must be able to accurately forecast market sales and moreover, their anticipated percentage of those sales. This paper addresses the former of these forecasts through the development of a multivariate time series model. Specifically, an autoregressive moving average (ARMA) time series model (otherwise known as the Box-Jenkins approach) is constructed using economic data relating to a 15-year period (1985–99). It is identified that population (millions); housing completions total (millions), total building repair and maintenance (billions Euro at 1998 prices, where 1 Euro = £0.6776) and gross domestic product (millions at market prices) are able to accurately predict plant sales. The performance statistics show the derived time series model to be very accurate in modelling the dependent variable [mean absolute deviation (MAD) = −87.04 and root mean square error (RMSE) = 1404.05]. By observing economic interactions and influences on sales, it is envisaged that the manufacturer can adjust their production output to match demand. Based upon the model produced, a forecast of machine sales for year 2000 is made and indicates that sales numbers look set to reduce to 15 482 units. The paper concludes with direction for future research work that will aim to model machine park, the influence of grey machines and individual plant items.
Forecasting UK construction plant sales
EDWARDS, D.J. (author) / NICHOLAS, J. (author) / SHARP, R. (author)
Engineering, Construction and Architectural Management ; 8 ; 171-176
2001-03-01
6 pages
Article (Journal)
Electronic Resource
English
Forecasting UK construction plant sales
| Online Contents | 2001
Forecasting Sales Volume in Construction Companies
| Springer Verlag | 2023
Forecasting automobile sales: the Pea-Box approach
| Online Contents | 2014
Forecasting automobile sales: the Peña-Box approach
| Taylor & Francis Verlag | 2014
Forecasting the sales of precast concrete building products
| British Library Online Contents | 1993