What makes and doesn't make a 'killer app' in civil engineering: a retrospective evaluation
Abstract: Every software developer, from the individual amateur to the largest enterprise, dreams of giving rise to a “killer application (commonly shortened to killer app) that is so useful or desirable that it proves the value of some under-lying technology”1. Whether the ‘killer app’ provides financial benefits either to the developer or the hardware platform vendor is beside the point. The important thing is the professional or social component: a true ‘killer app’ radically alters some form of human activity, either by creating an activity that did not exist before, or by im-proving the performance of an activity so dramatically that its practitioners view it as a revolutionary change. The first set of ‘killer apps’ so named, the early spread-sheet programs VisiCalc and Lotus 1-2-3, certainly revo-lutionized finance, accounting, engineering and many other professional disciplines. These programs, in fact, engendered the pursuit of the dream referred to above.
The title of the talk is not “How to create a killer app in Civil Engineering” but “What makes and doesn't make a killer app in Civil Engineering.” Forecasting is always a tough art. Given the wide range of human activities, it is even tougher to predict what tool will radically alter one such activity. Retrospective appraisal is much easier: you just need to evaluate what happened and attempt to trace from causes to consequences. Furthermore, because of the rarity of ‘killer apps’ generally, and in civil engineering particularly, it is not possible to treat the subject in any generic way; it can only be treated by evaluating exam-ples and attempting to generalize from them.
Keywords: Killer applications, civil engineering, structural analysis
Full text: content.pdf (202,755 bytes) (available to registered users only)
Weener R J
"The concept of hierarchical levels; an overall concept for a full automatic concrete design including the education of concrete. The case MatrixFrame? versus EuroCadCrete."
Abstract: "1. The exception proves the rule; Knowledge Based Automatic Concrete Design
From the early 80 till the second half of 1990 the software Matrix developed for structural engineers was based on the MsDOS platform. In those years the codechecking distinguished itself by an extremely enforced integrated approach. A complete structural design including the generation of drawings could be realized at once, with one press on the button. In this concept there was no room for the intervening interaction of civil engineers. They had little or no influence on unforeseen situations or shortcomings in the automatic analysis of boundary conditions or the automatic design.
The fact that we were secured of the cooperation of civil engineers (experts) concerning improvements makes it possible for us to make our knowledge based system even more complete.
2. The exception becomes the rule; Interactive Concrete Engineering
A disadvantage of a full automatic structural design is the existence of exceptional cases. Every case needs to be programmed which leads to a huge programming effort. In order to complete the last 20% you need a programming effort of 80% of the total period. Another disadvantage is the different approach by the government for using software for code checking.
The new Windows software is based on a structure very close related to the level of code checking. All the relevant parameters can be manipulated. The link to the code is absolutely clear by the visualization of the applied code article as well as the provided value and the required value.
3. The 80-20 rule; The concept of hierarchical levels
80% of his time a civil engineer is using only 20% of the functionality of his software for structural analysis. A program doesn’t need to be too complex for daily use. When you think in different levels you can manage the 80% for daily use, as well as the 20% for the advanced topics in 1 program.
The computer, using generative processes, without intervening interactions can work out 80% of all calculations. When you think in levels it is possible to work out the other 20% by the same program.
4. Ruling by exception; Computer Aided Learning system
10 years ago the TU-Delft developed a CAD exercise. During this period more than 1500 students used these exercises for their training. This CAD exercise was developed in order to support students in dimensioning, analyzing and detailing concrete structures, after the introductory lecture in designing and constructing concrete in their third academic year.
EUROCADCRETE is a continuation of the CAD exercise mentioned above and is based on the educational version of MatrixFrame 2D-Frame and on the experience of the TU-Delft during the lessons of the CAD Concrete exercise. Students at home can define the structural analysis part of the exercise. Then the prepared job can be worked out according to the EuroCode in the EUROCADCRETE environment. The last part of the exercise gives the student the opportunity to perform parametric studies. By means of exercises and by providing interactive tools students gain a clear insight in the nature of reinforced concrete, which is the aim of this job. A learning system like EUROCADCRETE is a combination of, on the one side, a Graphical User Interface based on the lowest level, and a check mechanism and parametric study on the other side, which is based on the advanced level within the concept of hierarchical levels."
Full text: content.pdf (877,357 bytes) (available to registered users only)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.man-software (0.062219)
Sound: read aloud.
Permission to reproduce these documents have been graciously provided by Icelandic Building Research Institute. The assistance of the editor, Mr. Gudni Gudnason, is gratefully appreciated