What is information technology in systems analysis


* Joachim Marks is a consulting economist and business economist in Kiel

Providing the right information at the right time is the central future task of information technology - according to Joachim Marks *, this goal has so far only been achieved by very few companies. The author creates a future concept according to which system and software development for the use of information as a resource can be carried out more effectively.

The software crisis has been invoked again and again - for at least ten years - and yet the demand for software services is growing all the time. In the meantime, hardware can only be sold with suitable software solutions. Visitors to the last CeBIT were able to get an idea of ‚Äč‚Äčthis development.

Should the frequently discussed thesis be correct, according to which the constantly increasing investments in software projects no longer pay off? If so, we may choose the wrong path into the so-called fifth Kondratieff. Because the upswing phases of the long economic waves, the so-called Kondratieff waves - named after the Russian economist who described these phases - are characterized by a particularly high return: Investments in these basic innovations pay off.

Euphoria is not

Proof of recovery

In the downturns, on the other hand, investments are made that bring below-average returns. This means that "too much of a good" is being done in these phases. In the exuberance of expectations of these innovations, investments are made which, given the level of factor prices - the prices for production factors such as machines, software or labor - no longer make sense, so that their profitability is no longer guaranteed.

Ultimately, this leads to excessive prices or shrinking profits (or losses), which can lead to entire branches of the economy being dragged into recession.

A general euphoria, as it was felt at CeBIT (see above), is of course no proof that we are really on the upswing of the fifth Kondratieff. The decisive factor is whether or not investments in information technology (IT) will generate above-average returns now or in the near future.

Are the productivity gains of information technology exhausted? Are bad investments now even to be expected due to the orientation towards old successes?

In my opinion, this is not the case, at least not in general, because we are only at the beginning of the development in the use of IT possibilities: both in terms of the worldwide linking of data as well as their linking within a company.

Information technology has only really led to new applications in a few areas.

Most IT solutions give the impression of simply shifting tasks that were previously carried out manually or mentally to a technical level.

A prerequisite for productive use of the resource information is that it is known at all what information exists in the company.

The aim of the information analysis is to make this transparent, as part of which a company-wide information model (data model) is created. Even if companies today have rarely created really complete information models, there is hardly any doubt about their usefulness. New projects usually take this aspect into account by creating partial models that grow together to form an overall model. The path to order the "data salad" is being taken. This is only the beginning, however, in order to fully exploit information technology and its possibilities.

Right information at the right time

It's not just about providing information; It is just as important to offer the right information at the right time - this is where the real qualitative advantage lies. All too often, information technology still stands alongside traditional organization today. The optimization of work processes - including the time factor - is often only marginally considered when developing new projects.

The systematic design of this process enables the so-called essential system analysis. On the basis of the events that trigger a system to react, that is, which cause it to become active, the model of a company (system) is created that focuses on the essence, on what is absolutely necessary even when using perfect technology. The essence of a system is limited to an explanation of what a system must do, regardless of how and by whom these tasks are carried out.

This view is based on the hypothetical assumption of a perfect technology. In this modeling step, it is assumed that the activities of the system are carried out at infinite speed, do not cause any costs and, moreover, are carried out without errors.

The system is therefore completely independent of the technology used. This original model can be used with technological advances, as well as with changes in factor prices that "force" a modification of the technology used.

What causes the (apparent) contradiction between the software crisis, which arises from the lack of profitability of software projects, and the increasing demand for software services?

1. The decisive performance factors of the structural change associated with the fifth Kondratieff, namely the quality and speed (time factor) of the production of goods or the range of services, are not or only insufficiently taken into account in the investment calculations.

2. The qualitative demands on the software generally increase the investment volume, so that the profitability of the applications depends very much on the number of individual products to which they are allocated. Quite a few company mergers should be seen against this background.

Two consequences for system development

Two consequences for system development can be drawn from these aspects: On the one hand, productivity must be measured more precisely. This means that measurement criteria for the qualitative and temporal aspects of the software products have to be created. In order to be able to draw up a complete investment calculation, the influence of the applications on the company's income statement must be estimated. It is only through such measures that it is possible to select the projects with the highest profitability and thus to control the investments sensibly.

The second consequence: The productivity of the system development has to be improved in a targeted manner. In general, it can be corrected from two sides, from the cost side, in that the production and maintenance costs of the software itself are reduced, and from the performance side, in that the productivity potential of the resource information in the applications is fully exploited.

The following measures are available for system development and software creation:

1. Measurable quality criteria are to be defined for the products of system development. Subjective assessments can also be used as a yardstick if objective measurements are not possible or if it would be too expensive to collect them. The product is to be understood here in the overall context of the organization and processing of a business area.

In addition to the quality of software and hardware, for example, the motivation of employees is also a criterion for determining the performance of a business area. The speed of processing should be considered as a special performance criterion, as this has generally become a decisive competitive criterion. Another well-known but important success factor is the estimation of the fixed and variable costs for the processing of the individual business transactions. If an explicit investment calculation is to be made, the influence of these criteria on the business success must be estimated.

2. Information modeling is a fundamental part of the development process. It structures the information and thereby creates a clear information situation, which ensures effective use of the information for potential actions of the "company" system that cannot be planned in advance.

3. Object orientation is to be included as a further principle of development. This increases the data consistency, reduces the maintenance costs of the machine systems and ensures the reusability of the processes.

For the future, development and maintenance costs for software development can be reduced.

4. If the productivity potential of the resource information is to be used optimally, the event-oriented essential system analysis must be a component of the software development. In this way, corporate systems can be developed that guarantee a holistic view of the company and thus integrate software and process organization.

Unnecessary dismemberments are avoided

The event-oriented entry summarizes the entire processing of a business transaction. Unnecessary artificial dismemberments are avoided. The process organization tends to be kept short. In this way, frictional losses, which can lead to increased processing costs or lost time, can be avoided. The type of approach is also easily accessible for the specialist department, so that communication problems with the developers are avoided. This leads to lower development costs and a higher quality of the product.

By explicitly creating an essential logical model, the basis is created from which completely new, effective, elegant and simple solutions for implementation concepts can be found that make grown, unproductive work processes superfluous or at least streamline them. In our experience, this applies in particular to secondary professional requirements.

The "big hit" is no longer in demand

These are requirements that do not directly serve the purpose of the system under investigation, but rather its management or control - for example, auditability, quality assurance or controlling. In addition, this step generates a result that can be used in the event of technological innovations or serious shifts in factor prices for the purpose of a redesign. This reduces maintenance and future development costs.

By assuming infinite speed in the essential model, the productivity factor time (speed) can be explicitly taken into account when designing a realization. The decision-making process for the selection of an optimal technological implementation can be controlled by the success factors of speed and unit costs of the business transaction.

5. Since the time factor is becoming increasingly important for the production of goods and services, including for system development itself, it is necessary to take suitable measures for the system development process to reduce the time between the start and implementation of projects to reduce. The "big hit" will no longer be in demand, but rather the continuous, step-by-step expansion of the systems.

In order to identify suitable sub-projects, a so-called top model is to be developed, which represents the events from the upper levels and the corresponding information objects. The model shows an overall system from which subsystems can be successively developed without creating isolated solutions. In addition to the individual productivity gains, additional synergy effects may be achieved through the link.

The measures for system development outlined above can be combined with one another and integrated into a procedural concept for system development. Components of such a concept are shown here to some extent:

1. Target definition of the projects, which leads to the derivation of the success factors of the system. These include quality criteria, the time factor and the variable and fixed costs of processing a business transaction.

2. Event-oriented essential system analysis with which TOP models can be created.

3. Structure of the essential functions according to the entities of the information model.

4. Selection of sub-projects that take place according to objective and target-oriented criteria.

5. Detailing of the information flow diagrams for the selected sub-area into a complete essential model In parallel, the entity-relation model is detailed and the detail functions are each focused on one entity.

6. Development of implementation solutions along the targeted goals and an evaluation based on the derived success factors.

As before, these first steps in system development must be followed by the phases of design, implementation, testing, etc. These phases should also be based on the success criteria. This is the only way to ensure effective, performance-oriented system development. The phases outlined above are not to be understood as a new waterfall model, but rather as partially repetitive activities in the course of system development.

The phases provide each other with input and control each other. So much information has to be gathered in each case that a decision can be made for the next steps.

In particular, alternative solutions must be taken into account. For example, a standard package or an organizational solution can adequately meet the goals of the project.

Summary: In order for information technology to be able to fulfill its decisive role as a basic innovation in the fifth Kondratieff cycle, the management of the system development must increasingly include the performance of the products in the planning. Only an integrated solution tailored to the needs of your customers can fully develop the productivity potential of the resource information.

Cost reduction alone is not enough

The discussion of the software crisis caused by information engineering focused too much on reducing the costs of software development. The performance of the organizational solutions produced will play a determining role that has been underestimated until now when it comes to productivity. As in the other goods and service markets, the price of the product is becoming less important as a competitive factor compared to new productivity factors such as the speed of availability and quality. The development strategy outlined here and in some cases already practiced takes these aspects into account systematically in order to synergetically strengthen the productivity effects of the individual methodological approaches.

Finally, it should be pointed out that it does not always make sense to exhaust the full canon of this development strategy. If the process-oriented success factors are of subordinate importance in the target catalog of a project and information management is in the foreground without a high number of cases, a purely object-oriented system development can be sufficient.

It should be noted that the choice of the methodical procedure can not only be made from the point of view of reducing the costs of software development, but must also take into account the efficiency of the organized systems. Only those who face market requirements in good time will be able to assert themselves. No competitive advantages can be achieved through cost reductions alone.