Guest Editorial: Reliability and Quality Case Studies

T HE real proof of Reliability and Quality Enhancement techniques comes during design, development, production, testing, and maintenance of products. Several papers discussing how such techniques have impacted products are included in this section, which contains five papers. The five case studies are from the United States, Japan, Canada, Hungary and Italy, respectively and discuss topics ranging from comprehensive programs covering all aspects of product life cycle to the impact 0f.a particular system component on reliability. The most comprehensive case study is given by W. Kremer, A. B. Sripad, and H. I. Saraidaridis in their discussion of a generic program being implemented on AT&T digital transmission products, as exemplified by the D5 Digital Terminal System. This comprehensive program includes consideration of quality and reliability in seven phases of the product cycle: Objectives and Budgeting, Reliability Studies, Hardware Reliability Enhancement, Software Quality Assurance, Qualification, Production Quality and Tracking. K. Tatekura, H. Yamamoto, H . Wakabayashi, and Y. Niiro discuss a system with very stringent reliability constraints, the OS-280M optical submarine repeater developed by Kokusai Denshin Denwa Co., Ltd., of Tokyo for the third transpacific cable system. The reliability target is not more than three ship repairs in the system life of 25 years, which corresponds to a failure rate of less than 16 FITS per regenerator for the planned system. A variety of techniques have been used to achieve these stringent reliability requirements, including choice of appropriate type of integrated circuits and duplication of transmitter paths, plus inspection, burn-in and testing of all components. The authors project that the selected components will meet the design goals; further, some componentsespecially the optical components-have improved in reliability remarkably due to their efforts. The procedures for fiber network design discussed by E. Roohy-Laleh, E. Abdou, J. Hopkins, and M. Wagner of Bell-Northern Research in Canada, have been motivated by distinct differences between fiber optic transmission systems and older copper networks. Economic deployment of fiber tends to result in structures with relatively low connectivity. Further, a fiber cable cut tends to, affect very nearly 100 percent of the circuits passing through the cable in comparison with, on the average, 56 percent of the pairs in a copper cable. Hence, a link cut in the most economical fiber optics network tends to be a major disaster in terms of network performance. The major techniques discussed are increasing the physical connectivity of the network plus use of a family of digital cross-connects and intelligent network control and operations systems to enable dynamic reconfiguration of circuits. Their techniques make it possible to achieve survivability for a fiber optic network greater than that of a comparable ‘existing copper network despite using significantly fewer physical links. G. Tarnai discusses the application of reliability and quality assurance techniques to train traffic control communications. He discusses a safety strategy developed at the University of Technology in Budapest for use by the Bulgarian Railways that focuses on elimination of failures, elimination of the consequences of failures and restriction of the consequences of failures. Methods and devices for functional tests of the equipment are discussed, with simulation of equipment’ functions and railway traffic having an important role in the tests. The final paper in this group, by F. Cucchietti, A. Gallesio, M. Liberatore, and A. Piccirillo discusses the impact of photodetector defects on performance of optical transmission systems installed in the Italian telephone network and operating at 8 and 34 Mbits/s. Under operational conditions, the measured bit-error rate (BER) turned out to be slightly greater than theoretical. values. When the BER was measured continuously over several months to identify a possible time dependent behavior, moreover, the error rate increase with time turned out to be much greater than predicted by theory. Tests to determine and correct this behavior are discussed. Each paper in this section treats a “success story” for use of reliability and quality assurance techniques during various stages of the life cycle of communication systems. Application of the techniques discussed has resulted in much more cost-effective systems than would have been developed otherwise. The need for consistent application of such techniques in the design of all communication systems of any real complexity or with a significant customer base is the main theme of this issue.