Parallel adoption

Parallel adoption is a method for transferring between a previous (IT) system to a target (IT) system in an organization. In order to reduce risk, the old and new system run simultaneously for some period of time after which, if the criteria for the new system are met, the old system is disabled. The process requires careful planning and control and a significant investment in labor hours.

Place in implementation process

There seem to be little conventions regarding the process of parallel adoption. Several sources (e.g.: Turban, 2002, Eason, 1988, Rooijmans, 2003, Brown, 1999), do not use a single process-description name. The term parallel adoption is denoted in these sources, although consistent per source as: parallel conversion, parallel running, shadow-running, parallel cutover and parallel implementation. This appears to be the case because a generic description of the process does not need a distinct classification. There are quite a few standard implementation methods, where different adoption techniques are described but often in a practical context; real-world case scenario or a more comprehensive set of implementation techniques like Regatta: adoption method, SIM and PRINCE2. In general, parallel adoption can best be seen as a systems engineering method of implementation of a new system. In principle, the parallel adoption method is different from the decision to change a system in an organization and can be seen as one possible mean to achieve that goal. However, there are quite some factors that are being taken into account in determining the best implementation strategy. Moreover, a successful implementation can depend to a big extent on the adoption method. (Lee, 2004)

Determining the parallel implementation strategy

The parallel adoption is preceded with determining the implementation strategy, which is not unique for parallel adoption, but can be seen as part of the change management process that an organization enters. (Lee, 2004). Some factors involved in determining an implementation strategy regarding adoption methods is described more thoroughly in Adoption (software implementation).

Risk versus costs

The reason for an organization to choose for parallel adoption in favour of a pilot conversion, big bang or phased adoption is often a trade-off between costs and risk (Andersson, Hanson, 2003). Parallel adoption the most expensive adoption method (Chng, Vathanopas, 2002, Microsoft, 2004, Anderson et al., 2003), because it demands from the organization that two systems run parallel for a certain period. Running two systems simultaneously means that an investment in Human Resources has to be made. Besides a good preparation of the (extra) personnel, that has to go through a stressful period of parallel running where procedures cross each other. (Rooijmans, 2003, Eason, 1988) Efforts should be placed on data-consistency and preventing data corruption between the two systems. (Chng et al. 2002, Yusuf, 2004 ) Not only for the conversion process itself, but also in training them for handling the new system. When it is necessary for the new system to be implemented following a big bang approach, the risk of failure is high (Lee, 2004). When the organization demands heavily on the old (legacy) system to be changed, the trade-off between extra involved costs for a less risky parallel approach, should be in favour of those extra costs (Lee, 2004), despite this, we see that ERP adoption follows a big bang adoption in most cases (Microsoft, 2004, Yusuf, 2004). This means that an organization should think clearly about their implementation strategy and integrate this decision in their Risk management or Change management analysis.

IT-requirements

To prepare the organization properly a requirements analysis of both IT-requirements as well as organizational requirements is necessary. More information on requirements analysis and change management can be found elsewhere. For parallel adoption, the most important IT requirement (if applicable) is attention for running the two systems simultaneously. In the conversion phase there is a timeslot, where the old system is the leading system. In order to transfer the data from the old system in the catch-up period to the new system, there must be a transition module available (Microsoft, 2004). Other implementation methods do not directly have this requirement. More information about IT requirements can be found in Software Engineering.

Preparing the organization

Requirements evaluation

The requirements evaluation involves redefining the implementation script. The IT and (if possible) organizational requirements that were made should be tested. Some tests can be run where the organizational responsibilities can be evaluated (Rooijmans, 2003) as well as the IT-requirements. Here it is also again important to have top-management support and involvement (Eason, 1988). If they do not make resources available to evaluate, the implementation can be unsuccessful as a direct consequence. After this evaluation the implementation script is redefined into a more explicit conversion scenario.

Conversion scenario

The conversion scenario thus consists of a blueprint for the organizational change in all aspects. However, there are two topics that did not yet get the attention they deserve in the parallel adoption scope. Workaround strategy / Rollback plan: Being distinct from the other adoption scenarios, also integrated in the conversion scenario is the workaround or contingency strategy with a rollback plan. The workaround strategy is defined in a broader scope in another entry, but in this context it indicates as defined in the above table: A backup plan; strategy taken on, in the conversion scenario to prevent errors in the conversion process and attempt to work around them, so that the implementation can still be successful. (Microsoft, 2004). The rollback plan, as being one possible workaround strategy, is initiated if something goes wrong in the conversion phase. Since the two systems run simultaneously, in a parallel adoption, the rollback plan indicates that the database or other system that handles the transactions should be fully retraceable in the legacy system (Microsoft, 2004). In fact the parallel adoption provides per definition this rollback plan due to its nature of a leading system and a (non-leading) backup system. Criteria indicators: Since the conversion scenario is a blueprint of executing the transfer of the two systems, is also entails quantifiable criteria. The redefined IT and organizational requirements are being transferred into measurable components. When the criteria are not being met in the test conversion, the workaround strategy should be deployed.

Conversion

The actual conversion phase is now in place. During this process, the organization is in a stressful period (Eason, 1988, Rooijmans, 2003). The two systems run parallel according to the conversion scenario and the new system is being monitored closely. When the criteria of the new system are met, the old system will cease being the leading system and the new system takes over. The catch ups that are part of the workaround strategy are the back ups of the old system and provide the means for reliability engineering and data recovery. There are two kinds of ways to make catch-ups: automatic catch ups and catch ups by hand. (Rooijmans, 2003). If applicable a remote backup service can be deployed as well.

Control system

Automatic catch ups: Catch ups that are being transferred by an automated system, created in the preparing the organization phase. This system automatically transfers the data or information to the new system when the conversion goes from the old leading system to the new leading system. The benefit of an automated system is that it is fast and accurate. The disadvantage is that is takes time to produce a transfer system in an earlier stage. Catch ups by hand: When the actual conversion entails only a small amount of time, or the complexity of information that should be transferred to the new system is small, an organization can choose to transfer the catch ups manually. The advantage of this procedure is that there is no need for a system (software program) to transfer the information and the possible problems that come with such kind of a transfer-program. The trade-off is accuracy and time. It takes a considerable amount of extra time, to transfer the catch ups manually and it is more vulnerable for small human errors (Rooijmans, 2003). Moreover, the additional investment in labour hours is high already; a manual catch up system places even more pressure on the personnel.

Evaluation / Practical relevance

There are several lessons that can be learned from case studies: The Nevada DMV system case, described by Lee (2004), learns that an implementation to a new process can also have a political implication. When the system that will be changed affects the general public and it is not only an internal system that is being changed, there are some more pressures that influence the organization. In this case, concepts as company image and reputation can drastically change if customers are faced with more delays in for example communication or ordering goods. It is suggested that if the system is politically sensitive, more attention should be paid to the conversion method and preferably parallel adoption is opted, since there is less risk involved. A series of lessons learned from a number of actual case scenario’s implementing a new portfolio system, performed by a business-consultancy firm (Venture, 2004) show some interesting lessons learned from the field. they seem to fit perfectly with the issues mentioned for a generic parallel adoption process, based on a combination of scientific work. To summarise: Risk assessment and contingency (workaround) planning is very important Assign project team roles Construct specific milestones (like PRINCE2) that include training and testing plans Identify potential risks and execute your contingency plan when necessary Communicate project status Changes should be appropriately authorized The conversion strategy needs to carefully examine the data requirements New and changed data should be tested against validation rules Construct a thorough rollback plan When possible, negotiate a pilot conversion There are also at least two difficulties with parallel conversion that may make its use impractical in the 21st century, though it was a staple of industry practice when inputs consisted of decks of punched cards or reels of tape. These are: 1. It is impractical to expect end users, be they customers, production line workers or nearly anyone else, to enter every transaction twice via different interfaces. 2. Timing differences between two multi-user interactive systems can properly produce different results even when both systems are operating correctly, are internally consistent, and could be used successfully by themselves. As a result, parallel conversion is restricted to a few specific situations today, such as accounting systems where absolute verifiability of results is mandatory, where users are all internal to the organization and understand this requirement, and where the order of activities cannot be allowed to affect the output. In practice, the pilot and phased conversion methods are more relevant today.

Articles

Andersson I. Hanson, K. (2003). Technology diffusion in a software organization, Licentiate Thesis in applied Information Technology, University of Goteborg Brown, C.V. & Vessey, I. (1999). ERP Implementation Approaches: Toward a Contingency Framework, Proceedings of the 20th International Conference on Information Systems, Charlotte, NC, December 13–15, 411-416. Chng, S., & Vathanophas V. (2002). Towards an Inter-Organizational Enterprise System: A Focus Group Study. The 6th Pacific Asia Conference on Information Systems (PACIS 2002). Tokyo, Japan. September 2–4, 2002. Lee, O. (2004). A Case Study of Nevada DMV system, Journal of the Academy of Business and Economics, Volume 3 Ribbers, P. & Schoo, K.C. (2002). Designing Complex Software Implementation Programs, 35th Annual Hawaii International Conference on System Sciences (HICSS'02), Volume 8 Yusuf, Y. & Gunasekaran, A. & Abthorpe M.S. (2004). Enterprise systems project implementation: A case study of ERP in Rolls-Royce. International Journal of Production Economics, 87, 251-266.

Books

Daft, R.L. (1998). Organizational theory and design. West: International Thomson Eason, K. (1988). "Chapter 9, Implementation and Support," in: Information Technology and Organizational Change. London: Taylor & Francis Turban, E. & Mclean, E. & Wetherbe J. (2002) “Chapter 14, Building information systems”, in: Information Technology for management. New York: John Wiley & Sons, inc Rooimans, R., Theye, M. de, & Koop, R. (2003). Regatta: ICT-implementaties als uitdaging voor een vier-met-stuurman. The Hague: Ten Hagen en Stam Uitgevers.

External links

Replatforming Line of business Applications from UNIX to Windows. (2004), version 1.0 Microsoft, Retrieved March 5, 2006 * Implementing a portfolio accounting system: Lessons from the trenches (2004), Venture Financial Systems Group Ltd, Retrieved March 6, 2006 [http://www.venturefsg.com/Lessons%20from%20the%20trenches%20-%20web%20v1.pdf

Text derived from https://en.wikipedia.org/w/index.php?oldid=1373074211 under CC BY-SA 4.0; modifications have been made.

Names in other languages