CONSTRUCTION OPERATIONS DATA ACQUISITION AND PROCESSING VIA TIME-LAPSE PHOTOGRAPHY INTERFACED TO A MINICOMPUTER.

This study deals with creation of a system for computerized data acquisition and reduction in the construction industry. An interactive computer system has been developed for collecting and statistically preprocessing data pertaining to construction operations into a form usable by available simulation and planning systems. The input to the system is frame-counts corresponding to initiation and termination of contruction activities extracted from time-lapse films. The inputs required for most analytical management tools in construction are activity duration times and cycles. The current method for extracting these duration times from time-lapse films is manual, tedious and time consuming. A comprehensive, economical and convenient method for data collection and reduction has not been available until now. This existing gap between the source of the raw data and the controlling systems motivated this research.

To develop the system the following steps have been taken: (1) An interface between a time-lapse projector and a PDP11/34 computer system has been developed to expedite and facilitate the process of acquiring the raw data from the time-lapse films. (2) The computer programs that form the system have been made interactive. Human factors considerations have been implemented in the design of the system. The intended users of the system are civil engineers and college students. The system has been made "forgiving" in case users err. (3) All the different elements of the system (computer routines) have been interconnected to form an integrated system. (4) A statistical package for analysis of the collected data has been converted for interactive use on the minicomputer. This package has been further modified to perform tests of hypotheses for the underlying statistical distributions of a data set.

The output data from the system is in the form of construction activity duration times. If adequate input data is provided, the statistical distribution of the data set can be estimated by use of the statistical package and statistical parameters of the population computed to be used in simulation models.