Fileread: Software Module for Reading Scenario Model Inputs and Observed Data from Text Files

摘要

SCENARIO is a computer simulation predicting thermal and cardiovascular responses to work in the heat. The present versions of SCENARIO enable model input data to be entered only by keyboard or mouse; there are no mechanisms for obtaining data from electronic files. This limits the use of SCENARIO because manual entry of input data for running multiple simulations or for running simulations in which inputs change many times (i.e., simulations of field and laboratory studies) are very tedious and time consuming. This also precludes the use of parameter optimization to simulate the responses of an individual, which requires direct access to observed data. Comparing model predictions (which are saved in ASCII text files) with observed data (saved in other electronic files) during model development and validation is also extremely tedious. The FILEREAD module described in this report attempts to alleviate these problems. Many simulations can be run in a batch mode through the use of a simulation specification file. For a given simulation, FILEREAD reads model input and observed data that are stored in ASCII tables. Any number of tables may be used to hold the simulation data. The study, test and subject identifiers are defined as key fields for querying records so that more than one study, test, and/or subject may be included in the same file. FILEREAD merges data from more than one file based on date and time fields, which must be sequential within each file for a given study/test/subject combination. Variables may reside in any order in any of the data tables; extraneous variables in tables are ignored. Variables may be specified in a wide variety of units. Each time FILEREAD is called from SCENARIO, model input data describing a single scene or set of input conditions is passed from FILEREAD to SCENARIO. This Technical note serves the purpose of documenting the FILEREAD algorithm and software module.