The Networked Enterprise

17 The Networked Enterprise Hubert Österle Networking is the lifeblood of the real-time enterprise. This chapter Hubert Österle describes where the networked enterprise is going and the roadblocks to Hubert Österle is director of the Institute of Information Management getting there. Before we examine the dynamics of today’s networked of the University of St. Gallen, where economy, it’s worthwhile to trace the evolution of the real-time enterprise he has taught since 1980. A frequent to get a deeper understanding of the underlying mechanisms. speaker and author on information management, business engineering, and process and system develop- Before the Networked Enterprise ment, he is also founder and presi- dent of the board of directors of The In 1967, Peter Mertens published a book on intercompany collaboration Information Management Group and integration. In 1969, I was part of a group of students at the newly (IMG). founded University of Linz in Austria.We were mesmerized by Mertens´ vision of integrated data processing [Mertens 1966]. At the time, the concept was based on file exchange and batch processing. In 1974, as an employee of IBM Germany, I was asked whether a partic- ular software package for financial accounting might have a chance on the open market. The vision was that each employee and program in a company would have real-time access to data entered by any employee. The solution ran on a database, but due to performance issues would probably not be used for business. This solution was an application developed by Hasso Plattner and the other founders of SAP. In 1984, August-Wilhelm Scheer published a ground-breaking book [Scheer 1984]. He described a data model for an integrated manufacturing company, a model that influenced not only database integration research at universities but also the development in many corporations. In 1986, Getzner Textil AG, a leading European textile manufacturer with 800 employees, evaluated a new software solution for business processes. The goal was to enter data once and allow every employee and program to access real-time, consistent data. Getzner chose SAP® R/2®,although the related hosting and software costs were at the limit of the possible for that midsize enterprise. Years later, Getzner calls that decision a key to its business success as a textile mill in a high wage country. In 1989, the Institute for Business Information Systems at the University of St. Gallen in Switzerland founded the Electronic Market Competency PART 3 151 Center (Beat Schmid). Research focused on inter-company collaboration across the rising Internet. In 1995, as part of a research visit to SAP’s re- search lab in Silicon Valley, I had the opportunity to discuss inter-company processes with Hasso Plattner. Ecouraged by similar visions we created a business networking research area at the Institute and, at The Information Management Group, we started implementing these ideas through seamless processes in and among companies. In 1997, at SAP’s 25th anniversary celebration, Plattner declared that the company had been following the vision of real-time ever since its founding and that it would continue to do so. The Internet hype of the late 1990s highlighted the enormous potential for connecting companies. The bursting of the dot-com bubble showed that the wired economy couldn’t develop in a big bang, but rather incre- mentally, using sound business and economic decisions, which Plattner emphasized all along during the dot-com era. The networked company is a fascinating vision for technologists, but is it also a prudent business goal? What is networking and what can it bring to a company? Automation The first goal of computerization was automation, with computers replacing human labor for highly repetitive tasks. Using programs for billing, payroll, and warehouse management lowered personnel costs. The programs could read huge amounts of data and produced huge amounts of data. Data exchange was via punch cards and magnetic tapes. The user interface consisted of forms and printouts. Computers provided processing capacity, replacing manual labor. Visionaries such as Mertens saw the potential of electronic data exchange between programs and dreamed of a network of programs, but turning that vision into reality was difficult with the limited tools at the time. Database Integration Magnetic drives brought about a breakthrough, enabling real-time data exchange between programs via databases.At the same time, CRT terminals resulted in a quantum leap in usability.The prerequisites were in place to ensure that every program and every workstation had immediate access to 152 PART 3 all company data entered in the system, and that the data was consistent, providing, of course, that the programs used the same data model. Database integration was coming, but it took almost twenty years from the time of the first trials at the end of the 1960s to widespread use in the mid-1980s. Integrated applications reduced or eliminated duplicate entries. Using the system for receiving goods, for example, a company could not only enter the quantity stored, but also automatically update financial accounting information with the value of the quantity stored based on the prices in the supplier contract. Such integrated applications also accelerated processes. Accounting could immediately report on the value basis of the quantity stored and more quickly assemble a balance sheet. Product cost planning could access the current average delivered price in real-time. Getzner Textil AG was able to realize its vision of real-time order confirmation.When a customer requests a price and delivery date for an order of 5000 meters of fabrics, the employee can answer the question while the customer is still on the phone. SAP® R/3® provides access to the availability of raw materials and starting materials, delivery times for starting materials, capacities of weaving machines, current cost data, and finally the conditions for that specific customer. Getzner can lower costs and, more importantly, it provides faster service for customers, which is essential in the fast-paced textile world. This is a competitive advantage as long as the competition cannot match this service level. Once they can, this level of service becomes a requirement for all competitors to meet customer expectations. Integrated processes and the underlying applications require formalization and standardization. By formalization, I mean that procedures and work- flows, such as customer-specific rebates, must be described in such a way that they take all possibilities into account. Formalizing therefore means modeling business reality in data and program functions. Prerequisite for interaction of programs is not only having a formal description of the world. In addition, all involved systems must speak the same language, i.e., use the same model of the real world. PART 3 153 STANDARDS TIME Information technology’s essential goal is to model, or describe, the world or some piece of it: a document, the sound of a pop song, a complete business process. But only by adhering to standard ele- ments standard data formats, standard communications protocols, and standard interfaces these models can be shared effectively and without loss of meaning. Indeed, in a world where such notions as collaboration, partnering, and integration are increasingly important, so are all forms of IT standards. Österle sees standard data and message formats helping companies to integrate their applications and processes across the Web. In Chapter 4, Scheer argues that standardized data models are crucial to helping decentralized enterprises make sense of their disparate databases. Along similar lines, Kumar writes in Chapter 15 that the push for real-time IT will inevitably drive standardization within companies and across the IT landscape. Further discussion of the importance of standards shows up in Kumar’s look at the growing use of outsourcing. In the 1970s and 1980s, many companies invested in companywide data models designed to create company-specific formalization and standard- ization. Unfortunately, the resources required to construct a prototype, particularly for realization of a companywide application system on the one hand, and political barriers on the other, impeded the realization of this vision. It took independent software houses, such as SAP, Baan, and J.D.