Testing software types

Whether you have one procedure or twenty to address this requirement you need to cover all types of equipment as well as test software if you use it. Some of the requirements of this section may need to be addressed in procedures developed for other purposes, such as the requirement for determination of measurements. 

Use is controlled by specifying the software by type designation and version in the development and production test procedures or a register which relates products to the test software which has to be used to verify its acceptability. You should also provide procedures for running the software on the host computer or automatic test equipment. They may of course be menu driven from a display screen and keyboard rather than paper procedures. 

The operational qualification for standard instruments, microcontrollers, and smart instrumentation consists of a black box test. This type of test is based on the user s firm application requirement and challenges a program s external influences. It views the software as a black box concerned with program inputs and its corresponding outputs. The black box testing must consider not only the expected (normal) inputs, but also unexpected inputs. Black box testing is discussed in Chapter 9. 

The operational qualification for standard software packages consists of a black box test. This type of test is based on the user s firm application requirements specification deliverable and challenges the packages external influences. 

A unit test is a white box-oriented test, conducted on modules that have been structurally tested. This type of test involves the standalone software unit before it is integrated into a complete system or subsystem. The test cases are designed to look for errors in the logic design, data processing, and the statement of execution. Uncovered errors are removed, and the test case that located the error, and any other associated test case, re-executed. 

The parametric method is much more complicated than the simple nonparametric method and requires computer software. The method is presented here under separate headings for testing of type of distribution, transformation of data, and the estimation of percentiles and their confidence intervals. 

IV.60. Documented measures should ensure that tools, gauges, instruments, test software and other inspection, measuring and test equipment, and other devices used in determining conformity to acceptance criteria, are of the proper range, type, accuracy and precision. They should be properly handled and stored, controlled, calibrated and adjusted at specified intervals to maintain accuracy. Records of calibration should be maintained and be adequate for traceability of measurement, to national or international standards, when necessary. When deviations beyond prescribed limits are detected, an evaluation should be made of the validity of previous measurements and tests, and acceptance of tested items reassessed. 

The examples given are for documents. Types of destructive testing carried out on production software products (typically described as deliverables) are well known and include module, integration and system testing. Specialized types of testing are common in specific environments, but all should have the same focus - cost of failure. 

The focus for the selection of test techniques is mainly on the parameters of effectiveness, defect type and software type and not on other parameters such as the modelling paradigm of the subject to be modelled. 

Aerospace struetwes made of composite. As part of the evaluation of the developed ultrasonic spectroscopy system the NSC software was tested on ultrasonic resonance spectra from composite panel samples. Spectra were collected with four different types of damages, and from flawless samples. The damages included a small cut in one of the carbon fiber... 

The Calibration of the positioning system is carried out using a bar with a given distance which is placed between the referenspoint on the microphone collar and the probe. The distance is then entered into the acquisition software together with informations of the air temperature close to the tested object, pipe dimension, type of UT-probe (probe height) and scanning direction. 

This type of scanning produces positioning data which is false in both x- and y-direction caused by the fact that acqusition software operates assuming that the shape of the test object is cylindrical. 

The assessment of the contribution of a product to the fire severity and the resulting hazard to people and property combines appropriate product flammabihty data, descriptions of the building and occupants, and computer software that includes the dynamics and chemistry of fires. This type of assessment offers benefits not available from stand-alone test methods quantitative appraisal of the incremental impact on fire safety of changes in a product appraisal of the use of a given material in a number of products and appraisal of the differing impacts of a product in different buildings and occupancies. One method, HAZARD I (11), has been used to determine that several commonly used fire-retardant—polymer systems reduced the overall fire hazard compared to similar nonfire retarded formulations (12). 

Neither database generally offers the possibility of integrating into it the greater number of values and test data that may already be established by users or processors. These organizations have data for their own internal use, and their goal has been to integrate all these types of data sources. Such in-house databases are at present available under operating system BS 2000 and in conjunction with the database software known as Adabas. 

Overall control is provided by the PDP-11/44, running DEC S RSX-llM operating system. RSX-llM is a multi-user multi-task operating system, and a number of other analytical instruments are interfaced to this computer system and are running concurrently. The automated Instron software is menu-driven because our experience has shown that menu-driven software is particularly effective for applications of this type. To perform either test the user accesses a main menu from which separate menus for instrument calibration, tensile tests, and flexure tests can be reached. The tensile and flexure menus have equivalent options the choices pertaining to automated testing are as follows ... 

The basic wood densities (dry) for different species were obtained from Ref. [36]. A basic density value obtained from the weighted average of the densities of each site s species was used for the species that for various reasons could not be identified. For estimation of SOC (soil organic carbon), equation (6) was used [30]. For data analysis, the nonparametric type test was chosen. We used the INFOSTAT software, and a value of 0.05 was considered significant. 

The computerized systems, both hardware and software, that form part of the GLP study should comply with the requirements of the principles of GLP. This relates to the development, validation, operation and maintenance of the system. Validation means that tests have been carried out to demonstrate that the system is fit for its intended purpose. Like any other validation, this will be the use of objective evidence to confirm that the pre-set requirements for the system have been met. There will be a number of different types of computer system, ranging from personal computers and programmable analytical instruments to a laboratory information management system (LIMS). The extent of validation depends on the impact the system has on product quality, safety and record integrity. A risk-based approach can be used to assess the extent of validation required, focusing effort on critical areas. A computerized analytical system in a QC laboratory requires full validation (equipment qualification) with clear boundaries set on its range of operation because this has a high... 

Cabrera et al. [50] modeled a set of 163 drugs using TOPS-MODE descriptors with a linear discriminant model to predict p-glycoprotein efflux. Model accuracy was 81% for the training set and 77.5% for a validation set of 40 molecules. A "combinatorial QSAR" approach was used by de Lima et al. [51] to test multiple model types (kNN, decision tree, binary QSAR, SVM) with multiple descriptor sets from various software packages (MolconnZ, Atom Pair, VoSurf, MOE) for the prediction of p-glycoprotein substrates for a dataset of 192 molecules. Best overall performance on a test set of 51 molecules was achieved with an SVM and AP or VolSurf descriptors (81% accuracy each). 

Statistical Analysis. Analysis of variance (ANOVA) of toxicity data was conducted using SAS/STAT software (version 8.2 SAS Institute, Cary, NC). All toxicity data were transformed (square root, log, or rank) before ANOVA. Comparisons among multiple treatment means were made by Fisher s LSD procedure, and differences between individual treatments and controls were determined by one-tailed Dunnett s or Wilcoxon tests. Statements of statistical significance refer to a probability of type 1 error of 5% or less (p s 0.05). Median lethal concentrations (LCjq) were determined by the Trimmed Spearman-Karber method using TOXSTAT software (version 3.5 Lincoln Software Associates, Bisbee, AZ). 

---