Network Information Systems for the Communications Industry

Network Information Systems for the Communications Industry

Richard Spooner
Marketing Manager NIS, Unisys Communications
Bakers Court, Bakers Road, Uxbridge
UB8 lRG UK

Abstract
This paper explores the competitive pressures which are behind the growing interest of communications companies in applying spatial information technologies beyond the traditional domain of ‘plan-build-record’. It argues that the need to compete on cost, quality of service, and time to market, are making it mandatory for operators to automate and integrate their business processes and systems, and it argues that this is creating both opportunities and challenges for spatial information technologies. It draws upon conceptual models from the fields of telecommunications and information systems, to describe how such technologies can assume a key operational or strategic status when implemented as part of shared model of network infrastructure - a Network Information System (NIS). It suggests that increasingly, the real value of NIS will have far less to do with expanding the network, and much more to do with optimizing performance in use.

The Customer ‘versus’ the Network
Competition - real and imminent - has been responsible for a major shift in the business focus of communications companies. They are being transformed from network-centric operations run by engineering, to customer-centric service organisations. In the course of this transformation, operators have tended to overlook many of the failings of their existing operational support systems for managing the network. Today few operators have in place a system for managing network information which is comparable in scale or sophistication with their customer information system (CIS).

The reality is that in many communications companies, the same networks will be represented in dozens of different data models, each associated with different operational support systems (0SS) for various activities which include network planning, provisioning, network surveillance, outage management, works management, and so on. There are often many manual transcriptions of information involved in the handovers from one 0SS to another.

Innovative communications companies are showing interest in implementing a shared data model of information on the network - a Network Information System - which offers the prospect of improving the flow-through of information and reducing transcriptions and handovers [1]. This increasing demand for NIS can be traced to the competitive pressures faced by modern communications companies, which we now discuss.

The Systems Integration Challenge
Behind the mergers and acquisitions which characterise the communications industry, there is massive internal restructuring. Adams and Willetts talk of a shake-out in the communications industry, in which only the ‘Lean Communications Providers’ will survive [2]. These will be the operators which have automated the end-to-end flow through of processes involved in service management and network management - much along the lines of the automobile industry. It is only through automation, the authors argue, that operators will achieve the reductions in cost, improvement in service quality, and reductions in time to market that will ensure their survival.

They identify three components to automation and process flow-through:

automating the process steps that were done by people, to be done by computer
integrating the process steps to reduce manual handovers of information
rebuilding processes to reflect new ways of doing business, as opposed to mechanizing existing processes

While the spatial information community has long argued that the greatest benefits of the technology derive from integrating and rebuilding processes, the reality in the communications industry is that operators have tended to limit the application of the technology to automating discrete process steps, and particular those steps involved in the ‘plan-build-record’ activities of network evolution. Many operators are now contemplating exploiting spatial technology beyond ‘plan-build-record’. They have evaluated the possibility of building on existing systems, and found in many cases that legacy systems for network planning have been put under stress by the need to compete. We can examine some of these pressures.

Network expansion and the role of spatial information
Networks need to be planned and designed, whether or not the operator is facing competition. However, the arrival of competition generally changes the planning process, and this in turn creates new challenges and opportunities for spatial information systems which have traditionally been so central to network planning.

For example, under a competitive environment, the pace of change is no longer determined by engineering requirements, but is customer- and market-driven. In the past, it was sufficient to base forecasts of demand on demographic and economic data for large areas, and extrapolate from historic patterns. Increasingly the business requires much more sensitive forecasts (in time and space) based on a variety of information sources (internal and external). New technology is being applied to specific parts of the network on an as-needed basis, rather than rolled out wholesale through the operator’s territory, according to an engineering schedule.

Planning has to be both proactive and reactive. The planner no longer has the unidimensional task of minimizing capital costs, but is expected to consider recurrent costs, and multiple objectives. He or she is increasingly expected to be multi-skilled, and able to plan networks based on pooled information rather than local knowledge. All these changes brought about by competition are challenging spatial information technology to justify its contribution to the business. It has to demonstrate that it can model multiple networks, that it can integrate the customer with the network, accommodate market intelligence, scale to hundreds of users, and support long and short transactions. In Adarns and Willetts terms, it has to demonstrate its capability not only to automate the steps of an existing process, but also to integrate steps, and to accommodate new processes. It is expected to demonstrate that it is more than a ‘supporting’ technology, and that it has the potential to support the company’s core business.

Integration and extension
A useful framework for positioning the strategic value of information systems in general, is provided by Ward and Griffiths [3]. It proves especially useful as a barometer of the extent to which an organisation has implemented a Network Information System. It is formed out of the intersection of two key dimensions of information systems - the degree to which the system integrates with other systems and processes, and the degree to which it can accommodate new information. The resultant grid categorises information systems as strategic, high potential, key operational, or support, depending on the current or expected contribution of the system to the business success of the operator (see Figure 1).

Figure 1 Categories of information systems (from Ward and Griffiths, 1996)

Many spatial information systems used in communications companies exhibit the characteristics of support systems. They were developed as standalone systems for the network planner or record keeper. They have proven valuable to the business, but are rarely critical for its success. The emphasis is on achieving productivity improvements which can be translated to cost reduction.

According to the model, the more significant business benefits derive from information systems which are located towards the upper right area - key operational systems and strategic systems. Key operational systems are systems which show a high level of integration, and which the company depends on for its current business success.

Communications companies using spatial information technology in a key operational system have applied the technology not only to the production and maintenance of network records, but also to the automation and integration of the business activities which use those records - asset management, costing, maintenance, work management, network monitoring, and so on.

For such a company, the emphasis is on using database technology as a repository for network information, and as means for sharing information between multiple users and multiple applications. The approach is data-driven rather than graphics- or application-driven, which are important characteristics of an NIS approach.

An organisation using NIS as part of a strategic system has extended the management of its network beyond current information, to encompass the management of new information - information which will be critical to sustaining its future business strategy. This new information can include information on new networks, technologies, and equipment; plus any information which is complementary to the network. This might include information from the CIS on the services being delivered, or from marketing on the potential demand for new services.

The challenge for many operators is to implement NIS as key operational and strategic systems. Many new entrants are setting out to achieve this directly, unconstrained by legacy systems and databases. Incumbents on the other hand, often have to deal with legacy support systems. A typical incumbent’s response to a new business requirement, for example to plan and manage a new cable technology, might be to develop a new application as rapidly as possible by deliberately avoiding integration with other systems. Many operators have built such high potential systems as technology-driven, IS/IT initiatives. These systems have the potential to bring significant benefits to a business, but have to be incorporated into the mainstream in order to realise this potential.

It is clear that the task of implementing an NIS is likely to be more difficult for an existing operator than for a new entrant who is less encumbered by legacy systems and legacy infrastructure. There are grounds however, for arguing that the increased difficulty and expense, is more than outweighed by the advantages. An existing operator’s greatest assets are probably its customer base, and the network infrastructure which supports that customer base. Incumbent operators have the most to gain by employing NIS to manage and operate the existing infrastructure, and the most to lose by failing to ensure adequate standards of network performance, as we argue below.

Spatial Information Technology and Network Performance
The monitoring of performance at both service and network level, and in all its manifestations - expected, required, agreed, designed, achieved, and perceived - has become a critical activity for any competitive operator. For the purposes of this discussion it is sufficient to restrict our attention to the performance of the physical network infrastructure alone. We will refer to a simple but very effective model proposed by John Mellis of BT, as an aid to understanding the spatial issues relating to performance modelling, as they affect the planning and operation of the network infrastructure [4].

Figure 2 Network infrastructure performance and planning (from Mellis, 1996)

Mellis’ model represents network performance issues as the intersection of three domains of the network: cost, transmission performance, and reliability. (Figure 2). The intersection of these 3 domains describes infrastructure planning activities which have traditionally been addressed by different systems, often each with their own model of the network. Mellis points out that developments in database technology, networking, and the availability of inexpensive computing power, have removed some of the technical obstacles to operators applying a more holistic approach to network planning.

As a general observation, spatial information technology has probably been used most successfidly by the network designer, in modelling transmission performance and costing different geographic and topologic designs. The technology has proven well-suited as an environment in which the designer can explore the effects of changing network designs on location, length and composition of plant, upon transmission loss, and the cost of materials and installation.

Spatial information technology has been least successfld in modelling the third component of the performance triangle - that of network reliabili~, which is so crucial to the customer and service focus of the modern operator. To model network reliability places high demands on the availability of accurate records on the placement and characteristics of network elements, and their condition - information which the operator may not have. But even if this information were available, would it be sufficient to allow us to model reliability ?

Mellis points to the considerable evidence from BT experience, that the reliability of plant is related not only to plant condition, but also to the quality of manual intervention in the network. This evidence leads Mellis to reflect on the truism that “. network performance at all levels is fundamentally affected by human factors. The speed and quality of maintenance, fault location, defect repair, planning decisions, and installation practices are critical influences, particularly for the external network, where these process issues may dominate the ‘designed’ performance of the plant.”

Unless the experience of BT in the UK is particularly unique, this suggests a major challenge for the application of spatial information technologies in communications. The way the network is operated ultimately affects the performance of the network, which we know to be a contributory factor in determining overall quality of service, and ultimately customer satisfaction. It might also suggest that in looking for the ‘big benefits’ from applying spatial information technology in the communications industry, we have been driven by the wrong paradigm - one which owes its origins to design- rather than action-orientation.

NIS to Support ‘Business-as-Usual’
There is undoubtedly a growing interest amongst the far-sighted operators to use Network Information Systems, to support what Mellis refers to as ‘business-as-usual’ activities. These include fault-finding, trouble-ticketing, fault repair, preventative maintenance, plant and premises locating, work management, job scheduling, complaint handling, and order processing. It is true that spatial information technologies have been used in all of these areas, for many years. However, what is new is a recognition of the value of integration which derives from sharing common models of the network infrastructure amongst multiple applications and multiple users. It is the difference between support and high potential systems on the one hand, and key operational and strategic systems. It is a trend which is evident in recent developments to provide end-to-end infrastructure management by integrating inside- and outside plant management systems. The trend manifests itself in the integration of spatial information systems with Network Management Systems, which are used for real-time monitoring, interpretation and control of the network. The combination of technologies provides an operator with a physical view of the network and a logical view of the network on one platform. Process automation and flow-through can be further enhanced by integrating with trouble-ticketing, works management systems, and vehicle dispatch systems.

The communications industry has some way to go to catch up with the utilities in making digital information on the network and customers available to engineers and salespeople in the field. The same is probably true of the use of spatial information technology in customer contact systems. By making network infrastructure information available to customer contact staff, operators can give rapid and accurate responses to customers over questions concerning provisioning of new services, and fault reporting. There is also untapped potential in using spatial information technology as a means to take a more pro-active stance on network maintenance. By integrating with line-testing and fault identification systems, NIS technology has the capacity to reduce the industry’s unusual reliance on the customer as the prime mechanism for fault reporting.

To be accepted as a key solution in telecommunications, spatial information technology needs to make a demonstrable contribution to an operator’s business success. In many cases this means going beyond the domain of pre-service activities, and embracing both future service and especially in-service phases of the network and services lifecycle. By addressing ‘business-as-usual’ activities, the technology is gaining acceptance at both Network and Service Management levels of the business. However, unless these systems are supported by a common model of the network, the operator can easily fail to realise the full potential of spatial information technology, which is not only to automate discrete processes, but also to integrate and change processes.

The ultimate challenge for NIS remains, as ever, that as an operational system, NIS makes far greater demands on the accuracy, availability and machine-readability of network records, than it ever did as a network planning and design tool.

References

Spooner R, The Case for Network Information Systems in the Communications Industry, Unisys White Paper,4134 9796-000, 1997
Adams E K & Willetts K, The Lean Communications Provider, McGraw-Hill, 1996
Ward J & Griffiths P, Strategic Planning for Information Systems, John Wiley and Sons, 2“dedition, 1996
Mellis J, “Systematic approaches to network infrastructure planning and performance modeling”, BT Technol J Vol 14 No 2 April 1996, pp87-94