Data migration is a very important part of any new system implementation and experts would agree that an unplanned migration will lead to failure to complete the project on time and in budget. There are a number of challenges.

Popular legacy applications were developed years (decades) ago so that many of those who supported the applications, understood the data structure and knew how to extract from it, are no longer in the business. There are also a lot of custom and home grown applications still in use to deal with limitations in legacy applications or with changes that have happened in the industry. And there is a lot of non-system data that the new system will require but that is currently on paper, in PDFs or other off-system applications. A due diligence exercise will usually reveal more data to be migrated than that in the legacy system alone.

Vendors might well have the functionality and the process but the user has to provide the right clean data… as the popular IT adage puts it: ‘Garbage In, Garbage Out’. No matter how many millions of dollars a new system has cost, if the data going in is garbage, there will be no return on that investment: the new system will look good but will probably function worse than the old system. There must be a plan to address data migration challenges. In this, one important challenge is the ‘extraction’ of data from legacy systems and other sources (see above) which requires a range of talents to convert them to a standard format from which they can be transferred.

At Hexaware, we have developed tools and processes to manage extraction and ensure that all the data required for the new system is first captured into a Staging Database. A very important component in the migration is how the source structure is mapped against the target structure; it’s not a case of simply moving data from ‘A’ to ‘B’. Each field on the target structure needs to have identified a source structure, transforms and rules to be applied, along with cleansing and integration methods. It needs to involve the migration team, the business team and the new product vendors. Once the mapping is complete, the transfer can proceed into the new target structure and the files required by the target database.

Even at this stage, there will be a number of checks to be applied as the data is transferred. In a nutshell, if we look at a high level, we can see the different stages of data migration…

This is a complex process that will benefit from meticulous planning and a diligent completion of all the stages illustrated above. Over many years, we have developed a number of tools and processes to support the various stages of migration from legacy systems to the new generation of MRO systems. Our experience has included migrating more than 500 aircraft on more than ten systems involving more than 20 technology skills.

Case Study
To illustrate what is possible, we have recently completed a migration from two existing MRO systems (one for maintenance engineering and one for materials management) into AMOS for a large EMEA airline, Part 145 certified, with a fleet of more than 150 aircraft and with ten years of historical data to transfer. The plan was for an 18 month migration and that was achieved despite encountering many data quality issues and data was in multiple formats. We were also able to ensure that the data integration and quality in the target application were good enough to leverage all the functionalities of the new system, and ensure a return on investment.

Data warehouse and analytics
Now, having migrated to a new system, what additional value can be realized for MRO? One key subject area of interest is in analytics. There is no doubt that it will shape the future of our business. Some product vendors and OEMs are developing a raft of operational analytics around the data generated by aircraft which will prove very useful for the industry especially in securing a safe future. But we still need to have an enterprise-wide consolidation of pure data to support analytics throughout the business to generate a competitive advantage.

One airline with which I have worked has recently undertaken an initiative to optimize inventory with the objective of reducing that inventory by $100million. This is a common objective but how can it be achieved? It is first important to understand the data around the inventory – what are fast moving and slow moving items; what are critical items; how is supplier performance around these items? Again, there are challenges, especially the integration of data from different systems, and also opportunities that can be derived from the analytics platform when it’s applied as part of a proactive maintenance program.

An important aspect of predictive maintenance analytics and reporting is to ensure that the proper foundation is installed. There can be good looking reports and dashboards at the front end but if the back end of the system cannot provide the right data, it is not a value adding report or dashboard. So it is important to build the foundation well, looking at the different sources of data (see above) and using an industry standard data model in which to amalgamate all of the data to supply the required reports.

Once a good data warehouse has been built, the dashboards and reports can all be built around it. But there is a choice whether to build the system from scratch or to apply a packaged solution. One of the first advantages of a packaged solution is that it is already in existence and so potential users can see all of its capabilities from the outset – there’s no need to wait for them to be created. Packages are scalable, can be customized, can accommodate additional KPIs and can leverage any Enterprise platform. They can also be deployed quickly, in weeks or months as opposed to the quarters or years it can take to build a solution from scratch. Total cost of ownership will be lower because individual users don’t have to buy any licenses

While we’re looking at analytics and the future, it would be useful to also look at big data, the big thing of the moment. Opinions vary; some think that big data is not really useful for MRO at the moment while others believe that MRO is already late to the game as far as big data is concerned. We have tested some of the big data platforms and there is no doubt that the new platforms offer a significant advantage when processing the data. Of course, data volumes are increasing inexorably and exponentially with some aircraft generating gigabytes and terabytes of data. But, for that to be of any use, airline’s and MROs platforms need to be capable of processing those massive volumes of data.

Predictive maintenance
What all of the above is leading to is ‘what can we do with all of this data?’ For MRO, one important application is predictive maintenance, something of which any head of maintenance and engineering today will be dreaming.

There are many predictive analytics tools available and it will also require algorithms according to whether the system is predicting a failure or an event. Users will need to put in a lot of subject matter expertise. It’s not as simple as building a system and then getting a predictive output tomorrow. It requires a lot of hand-holding; the algorithms need to be fine-tuned and the system has to receive feedback in order to deliver the proper predictive output and, as has already been mentioned, a lot of data from different platforms will need to be consolidated.

E-enablement: mobility and integration
Once a proper platform has been established with proper analytics – the data warehouse has been built – what is the next step? How can all the actors around the business be enabled. New technologies offer many opportunities to enable those actors quickly and easily; those new technologies fall into the phenomenon SMAC (social mobility and cloud); delivering flexibility and integration, and data capture capabilities. But, no matter how good they might be, how can these platforms be leveraged to deliver all of the benefits that we want?

With the proliferation of devices, the possibilities are enormous and, with that, connectivity has improved so that the devices can be used anywhere and still be in the system. This improves speed and agility in doing business and saves time. There are a lot of solutions that can be built around e-enablement but users will want to first implement the ones that offer best return on investment (ROI). It’s also important to make sure that the data used is secure and that it can be accessed on different devices.

Mobility and cloud: emerging trends

Some of these (see illustration above) will be familiar to readers while some, like augmented reality, might be less familiar. Say a technician is walking to the aircraft and has a device with him which he can point to a particular aircraft. As soon as the device ‘looks’ at the aircraft it can tell what are the tasks pending for that aircraft and can go one step further to access the manuals for any components to be checked or repaired, including the ability to drill down to sub-component layers. Of course, we have to start with small steps but, because mobile enablement projects do not require heavy investments, a lot of small initiatives can be accomplished quickly and easily.

 

Jacob Baiju, Practice Head MRO IT Services, Hexaware Technologies
Jacob Baiju has over 15 years’ experience in IT consulting with the travel and transportation industry. In his many years with the industry he has worked in various roles including consultancy, program management and practice development. He now leads the MRO practice for Hexaware globally with a focus on providing innovative solution frameworks for the MRO Industry in key areas for their business transformation needs.

Hexaware
Hexaware’s MRO IT service is part of a Global Travel and Transportation practice with over 1200 consultants working with leading travel and transportation customers. Catering to three out of the world’s Top 10 airlines, Hexaware is one of the fastest growing IT services organizations. The MRO practice leverages its experience to create tools and services which help customers efficiently migrate to newer M&E Systems, implementing industry specific Data Warehouses and Analytics as well as Independent Application Testing Suites.