Thursday, 30 June 2011

The importance of disruption literacy with case study

There are three types of disruption events: unavoidable (things like the acts of nature or industrial strikes out of airline control), intentional (result of conscious risk taking like expanding operation at congested hub airports), and avoidable disruptions that are rarely recognised as such. Because they are not methodically separated, and therefore not well understood, avoidable disruptions are often described as a result of chance, allowing the oversights, errors, and ignorance to grow unnoticed and increasing the business risk. Ability to ‘read’ disruptions and identify avoidable ones can make a major shift in airline practices linked to improvement in operational and cost efficiency.

Each disruption event has a meaning and carries valuable messages to the management, operational and system planners. Understanding that it is a result of multiple interrelated activities, means going beyond subjective delay coding, false punctuality measures, and simplistic cost calculations and start talking to people involved in schedule recovery, operational and schedule planners, as well as network and fleet planners when necessary. Listening to their interpretations of events and letting them communicate on regular basis is the best way to promote knowledge about internal dependencies, and inspire new ideas about prevention or reduction of disruption risks. This valuable feedback is crucial for the understanding of true causes and wider consequences of disruptions. This process opens great opportunities for efficiency improvement in almost every part of an organisation.  Learning the technique of how to map the process and focus on things that matter most can make the introduction of these practices in everyday life even more rewarding.

The following case study of a disruption audit of a young low cost airline shows how the analysis of disruptions caused by a scheduling error can help reveal avoidable system weaknesses.

Disruption troubleshooting – case study

An airline was faced with an increasingly high level of disrupted operations. Senior managers were keen to do whatever it takes to decrease the number of operational changes that have obviously contributed to already high losses. The management team decided to carry out a pilot project, thoroughly examine three days of heavy disruptions, find out the volume of costs involved, and more importantly, get to the root causes of the problem beyond codes published in the delay report. The following example illustrates the scope of information that was obtained to evaluate management issues hidden behind these disruptions.

Event description
At the beginning of the winter schedule, several aircraft were pulled out of service by the maintenance department for ‘unscheduled maintenance’ without any previous notice. Disruptions were of a large scale and it took the airline six days to return to normal operations. The following information was obtained through the disruption information system, and was partly based on the reconstruction of events in order to reinstate missing data and erroneous inputs.

Disruption report
Number of disrupted flights
Direct causes of disruptions
61%  Unscheduled maintenance – 6 events caused changes on 35 sectors
38%  Late aircraft delivery – two events caused changes on 24 sectors
1%    Late crew boarding
4 - passenger numbers not available,
Subchartered flights
Crew positioning (flights/ground)
Aircraft positioning
Aircraft utilisation
98 aircraft hours out of service
4950 passengers experienced delays over 30min
2120 passengers experienced delays over 2h
593 passengers experienced delays over 8h
257 passengers from cancelled flights received compensation
Total disruption costs, 3 days
Cost breakdown
€124,200 Hired  replacement aircraft
€47,600 Additional technical, ferry and sub-positioning flights
€35,600 Cost of passenger delays
€24,100 Diversions (3 flights including positioning)
€14,600 Passenger compensations on cancelled flights
€7,400 Crew cost repositioning
€2,100 Other

In addition, a thorough root cause analysis was carried out by talking to people who were directly and indirectly involved in this ‘case’ – COO, Ground Operations, Operations Control, Maintenance Control, Crew Planning, Financial Services and Reservation – which took less than a week. The following are extracts from the report related to direct and indirect causes of disruptions.

Root cause analysis
After an in-depth analysis of factual information followed by reconstruction of events with operational decision makers and by information gathered from other operations managers, the focus was shifted from purely ‘technical’ reasons towards the less tangible system issues.

Management issues            Senior management are not much involved in resolving operational issues, apart from discussions at weekly delay meetings. There is a lack of interdepartmental coordination which affects operational results. Without much instruction about the priorities in decision making, the Operations Department is tacitly given the authority to make costly decisions, almost on daily basis. Responsibilities are not clearly defined. Staff members do not have a clear picture about who is doing what and how their decisions influence other processes. There is a lot of improvisation made by inexperienced staff, with much use of a trial and error approach to solve problems, sometimes with harsh operational consequences. A blame culture is preventing many of these things from being resolved. Many serious issues are not discussed with management. Responsibilities for schedule changes between the Scheduling and Operations departments are not clearly defined and this is reflected in the number and volume of disruptions. Supervision is rare and inconsistent. The management reporting system does not systematically provide enough of the information related to operational issues. Communication links are weak especially between Scheduling, Operations and Maintenance functions. There is almost no control over outsourced services including financial control.

Network and scheduling   Problems were revealed at several airports and routes, mainly relating to short turn times, late arrivals at airports with night curfews, and inappropriate block times. For example, problems with turnaround times on inbound flights from a major airport resulted in extremely low punctuality of 46 per cent, compared with 76 per cent on outbound sectors, with cascading effects on other flights.

Operations Control Centre (OCC) Lacking clear instructions from senior level, operations managers have set up internal priorities for schedule recovery, applied whenever the flight is more than four hours delayed. The first action on their ‘recovery list’ is flight cancellation, followed by aircraft rental. However, these policies are not always applied due to a fear of incurring high costs. Decisions are made with hesitation and by the time they are resolved, often by cancelling flights, new problems would occur, creating even bigger issues. There is a high number of hidden errors present almost in every area of operations and even schedule planning. As an example, numerous crew changes were the result of input errors and omissions that occurred during the frequent schedule changes. In addition, the accuracy of movement messages is questionable due to erroneous manual inputs. Information about delay reasons is often entered inaccurately and is not descriptive enough, which prevents analysis. There is lack of discipline in filling up voyage reports which affect the quality of operational information. In complex operational situations, OCC expects instructions from a higher level, while the higher level expects recommendations from OCC. These unresolved issues create additional pressure on operations controllers and are the causes of unnecessary disruptions. There is a problem in communication within the OCC department. People predominantly use email, even in the same department, and there are rare discussions about problems they need to resolve collectively. Problems with exchange of information between Operations and Ground Operations staff are ongoing where Ground Operations often complain that they do not receive on-time information on disruptions from OCC, which is the reason for their late actions.

Ground operations             The quality of service at outstations is poorly managed by third parties and is not properly supervised causing additional inconvenience to already stressed disrupted passengers. Control of invoicing by third parties at outstations is non-existent: hotel accommodation, meals, and transport costs during operational disruptions are fully controlled by agents without any feedback to the airline. Refund policy is defined but there are problems with its implementation. Procedures in the Airline Services Procedures Manual (ASP) are not being updated regularly, resulting in inconsistent services. The ASP Manual is not consistent with the ‘Conditions of Carriage’ used by the Call Centre. Ground Operations managers are unaware of the volume of irregularities at base airports and outstations and their effects on passengers and costs. Commercial priorities are not specified. Communication with technical bases (third-party service) is sporadic with insufficient supervision. Passenger compensation claims resolved after long delays. 

Operational information   Delay reports are regularly distributed to the top management team, but contain only essential, unreliable information. Cancelled and diverted flights are not included, nor the information about the number and costs of hired replacement aircraft. Operational data input is critical, as there are too many errors. Operations people are not fully trained to use the new operations control software.