The economics of fleet management computing

A little over a year ago, Los Angeles began developing its specifications for a new fleet management system.

It was at this point that the fleet department bumped into a hard reality: The city’s Information Systems (IS) infrastructure could not support client/server computing.

Not, that is, without significant added expense for upgrading hardware and software platforms as well as the city’s telecommunications network.

The result: The city’s fleet is now running on a new, state-of-the-art host-centric fleet management system. But there are no regrets, because the mainframe solution meets all its operational and functional needs just as well as the client/server system.

Plus, it plugs neatly into the existing IS infrastructure.

Los Angeles’s fleet department was at a technical crossroads. It’s aging, homegrown fleet management system was barely able to keep up with a fleet that had grown to more than 10,000 units.

Nor was the city able to link multiple repair facilities onto a centralized database, a step that would allow managers to analyze fleet data, compare performance over time and make informed decisions.

Today, there are no right or wrong answers in fleet computing – only sound economic choices to make.

The distributed processing capabilities of the client/server computing system are powerful, and as such, they are well suited for all city, state and municipal implementations. This is especially true if the necessary networks and hardware/software architecture are already in place or soon to be there.

On the other hand, the lower costs and proven durability of host-centric computing continue to make it a viable solution.

Five or 10 years ago, fleet managers had more latitude when specifying a fleet management system, and most small and mid-size governmental fleets typically opted for a departmental system.

Now these “stand-alone” systems are being rapidly replaced by highly decentralized fleet systems that link multiple facilities and departments onto a unified database. When client/server computing is added to the mix, there are additional factors to consider and additional expenses to incur.

From the outset, the fleet department should understand the capabilities and limitations of the organization’s IS infrastructure — now and into the future. If an organization’s strategic IS platform is mainframe based and the host computer has the capacity to handle a new fleet system, a host-centric solution has obvious economic advantages. Consequently, representatives from data processing, network management and telecommunications should be intimately involved in developing a specification for a new fleet management system from the earliest stages.

As a rule, highly decentralized host-centric computing is more cost effective than client/server computing, both in initial costs and in maintenance and overhead. One recent study, for example, found that over a five-year period, mainframe installations averaged three cents per transaction compared to 43 cents for PC/LAN installations.

In almost every instance, decentralized host-centric implementations require less expensive hardware and network software and are easier to install, therefore requiring less maintenance than comparable client/server implementations.

Yet, the distinct advantages of client/server computing can justify the higher costs, especially if the future strategic direction of the organization’s IS infrastructure is moving toward a fully integrated client/server environment.

End users particularly like the familiar Windows environment client/server computing supports, where all the programs have the same look and feel.

But it is the ability users have to switch easily between the various software applications running on their desktop computers, copying data from one program and pasting it directly into another, that sets client/server apart from 1980s host-centric computing. These enhanced capabilities encourage greater utilization of the fleet system – from the shop floor and inventory room to the offices of fleet and operational managers.

Advanced computer technologies are giving governmental fleet departments more alternatives to consider. There are clear benefits from both computing environments and have fleet management systems that will support either client/server or host-centric computing. Host-centric computing is not going away. If anything, the technology to support host-centric implementations is growing right along with client/server, with the potential to plug right into the Internet using dumb terminals featuring graphical user interfaces.