Rolls-Royce is distinct from other companies in that it shares the same standards that are applied to the development of its products – some of the highest quality in the world – with its IT.
But that’s not the only difference – the level of rigour for project management is also far removed from other organisations Jonathan Mitchell has worked at, he explains.
“There is a ruthlessness about capturing the right requirements,” he says. “We can take a year on requirements before we begin a project. An important aspect of this is to identify the correct process owners, and dissuade those who thought they were owners.”
And it is no coincidence that Mitchell’s full job title is director of business process improvement (BPI), as this is where the main thrust of his work lies. “The real key for us is process change, and IT is the enabler to this,” he says.
Rolls-Royce is one of the world’s most recognisable brands. It came into being when Henry Royce, who produced the first car in 1904, joined forces with Charles Rolls. The legendary six-cylinder Silver Ghost was launched in 1906 and in 1914, at the start of the World War I, Rolls moved into aero engines.
Today the company is a world leader in civil and defence aerospace, and marine power and energy, with an annual turnover of more than £7 billion. The emphasis on engineering excellence is evident throughout the company, from the precision-engineered engines it makes, to the way it manages its projects and business processes.
Attention to details, deadlines and budgets is typical for project management and is part of the corporate culture, according to Martin O’Dowd, project manager for theenterprise programme at Rolls-Royce
“During the last 10 years project management has been very good indeed, and is based on excellence.”
The last two years have seen Mitchell and his team concentrating on a massive enterprise-wide SAP implementation programme.
It began with the integration of SAP between the civil aerospace and energy divisions of the company, and has become a standard template throughout the organisation.
This has, of course, included moving all its processes and operations into a single standard. MRP, HR and financials all use the shared services model, and the global convergence programme has easily met its original completion deadline of the third quarter of this year.
The implementation was done unit-by-unit, and cost tens of millions of pounds. “We were looking for the business benefits, and a large number of facts independently affected this. One of the most important benefits was to release significant manufacturing capacity, and free up the supply chain,” says Mitchell.
In all, some 21 sites use the common template. The system covers sites in the US (Indianapolis, Mount Vernon, Washington, Houston), Canada, Singapore, Germany (Obereusel, Berlin, Arnstadt) and the
UK (Derby manufacture, Derby assembly and test, Derby repair and overhaul, Hucknell, Coventry, Bristol, East Kilbride, Inchinnan, Loughborough, Barnoldswick, Sunderland and Liverpool).
Further smaller sites will be added over the coming year. The template is also standardised across different business units. These include civil aerospace (which includes both large aircraft, for example, Boeing/Airbus and corporate jets), defence, the repair and overhaul business, naval marine, component manufacture and the new fuel cells business.
Improving capacity has become important to Rolls-Royce because its manufacturing workloads are increasing as the Trent 1000 and 900 engines (which power the Boeing 787 and Airbus A380) have come online at Derby.
O’Dowd comments: “There was a transition so that assembly of the V2500 engine for the Airbus A320 and the Tay engine for corporate jets could be moved to Berlin. This was startling and traditionally not done. It is rather a difficult process when there are 25,000 bits in an engine, and a couple of thousand part numbers. The assembly technology performance and quality process must be perfect.”
In recent years Rolls-Royce had not dealt in high numbers of engine units, but the demand for short-haul aircraft means that in six months the capacity has risen to around 1,000 units – a 20 per cent year-on-year rise since the post 9/11 trough.
The relieved capacity the single SAP instance has created means that work can be pushed out from single factories into a global network. “We think it is unique,” says Mitchell. “We can look across the whole of the supply chain, and we are starting to do large-scale forecasting and load balancing across the network.
“Process improvement is key, and we are using SAP in areas that were different,” he comments. “Now there are few customised systems, we operate vanilla computer systems and our processes are increasingly standardised. Upgrades are much quicker and easier to implement, and we feel the SAP implementation has been done well because it is based on business processes and not the technology.”
As is the case with most IT projects these days, Mitchell’s team had to include some unplanned work along the way. This included implementing systems for its joint ventures around the world. One of them was for N3, a joint venture with Lufthansa, which undertakes engine overhauls and OEM repair work. This became live in two stages in February and April, after work began in October last year. This was additional to the planned workload, but O’Dowd says the reality is that both in Derby and the joint venture, after training, the systems were deployed with very few issues.
“We are as close to vanilla SAP as we can be,” says Mitchell. “This is successful as we have common processes across the group. We would rather compromise and have 99 per cent, rather than have to customise to get 100 per cent.” Having a single system to work with means it is easier to carry out maintenance with the joint ventures, and the total care concept (see box), where there is less chance of counterfeit products entering the supply chain. “Control is at a much higher level, and we consequently have a much better ability to trace parts,” says Mitchell. “The SAP system allows more work around the network, and if there is an issue in one type of engine, changes can be made to all the models across the fleet. For our whole fleet of engines it means higher reliability going forward, which is what our total care model is all about. Our database now has excellent product data about the fleet, and it is much easier to find out what is happening.”
Mitchell formerly worked in the pharmaceutical industry and says there are many similarities between the businesses, with long product lead times, a highly regulated environment, and an emphasis on product safety. But there are differences too. “There is little in-house IT at Rolls-Royce, and a large outsourcing segment,” he says. The company’s main outsourcing partner is EDS, which has around 1,000 staff working at Rolls-Royce sites.
At the peak of the SAP implementation work he had 120 EDS people working alongside 300 from Rolls-Royce. “The blend of people was crucial,” he says. “The business knowledge came from Rolls-Royce and the right technical blend, from getting the right people from EDS. We found that EDS’ technical experience in SAP is very good, and we were able to use their specialists in the US and UK to provide a ‘follow the sun’ capability during the critical commissioning phase.”
Outsourcing is pushing the game for Rolls-Royce according to Mitchell. “We are upping what we require from our outsourcers all the time, but there has to be a business understanding. This project was led by the business, and the outsourcing specialists integrated into the team in an ‘extended enterprise’ model. The ERP team, withERP consultants, and engineers in the EDS team have to live it with you. It is a pact for responsibility, and there is an unworkable sense of detachment if the outsourcer is not physically there working with you.”
By last December the systems were complete, and the company carried out three full dress rehearsals to make sure the code was stable before moving ahead. “We scheduled a launch window which was carefully synchronised with business cycles,” explains Mitchell. There could be absolutely no compromise on testing. It is too important and for us it is a completely locked-in process. We’ve learned not to concertina the testing phase for our engine projects and we therefore will not do it for our computer projects either.”
Training at the company is stringent too. It’s MRP controllers are trained and competency-tested after which they are issued with licenses that are only validated for a certain period. When the time expires for critical roles, they undergo more training and further competency tests, and O’Dowd’s team reviews it every time there is a job change, as it helps to enforce a continuous rigour in all the business processes.
Although the cost of the project has been high, the benefits are higher, with effusive post-implementation reviews, and positive independent reviews throughout the company. “It has also allowed us to attempt larger more significant projects now, and to continue structuring and transforming the business around the outcomes,” says Mitchell. The COO is heavily involved as the sponsor and his focus on benefits drives a total team effort.”
Another project that Mitchell’s team did at the same time was updating its warehouse system. In the civilian aero engine business Rolls-Royce routinely warehouses bits of equipment, kits, spare parts and maintenance kit for aircraft worth around £60 million a week.
“We had legacy mainframes, and substantially needed to increase capacity,” says Mitchell. “This would mean increased efficiency through careful use of RF in the warehouse, but we had to be cautious, even though it would mean substantially increasing throughput.”
The project worked in parallel with the established systems, simulating outbound equipment over three weekends. “If we got it wrong it would be very difficult to recover and would have adversely affected the company’s financial position. We therefore ramped it up over time,” says Mitchell.
It was completed over Easter, and testing was rigorous once again, with dress rehearsals taking place three to four weeks ahead. “Sales order processing, dispatch etcetera. All these elements are important as if we lost of control of the system we could quickly reach an emergency situation – lots of people would notice and it could affect our market share. We did not want large numbers of airlines on the ground and countless angry passengers waiting for spare parts.”
Financial systems are now working well across the group, with European and North American finance centres all operating from a global purchasing perspective. Mitchell says there is a big difference at the lower end of purchasing, and it has also cut IT costs. IT investment now tends to be at around 2.4 per cent of sales at Rolls-Royce. Mitchell cites the standardisation policy as the reason why.
“Just about the only system customised is the predictive service. The benefits of working like this affect 80 per cent of the software. “We are different where we need to be, the same where we can be,”
The SAP standardisation programme is being hailed as a major success. It has used new builds and overhauls to provide a single template for the company: “It is an IT project that has run early, and the single operation is allowing us to open up our capacity and improve effectiveness. Everyone is pretty pleased with the results,” concludes Mitchell.
Total care concept
Traditionally, jet engine maintenance is carried out as a time and materials spares business. However, Rolls-Royce offers an alternative model based around ‘Total care’. In this model, the airline pays a subscription for every hour the aircraft is flying and in return Rolls-Royce and its joint venture maintenance units take care of all maintenance and overhaul activities.
To ensure the highest levels of availability, Rolls-Royce collects data from engines in flight and transmits it in real-time to its control centre in Derby. Emerging issues can then be dealt with before they cause disruption to the airline’s schedule. Large-scale data mining is also undertaken to find trends that can be used to achieve even higher levels of reliability.