|
|
Optimising Flow
|
|
|
Optimising flow through today's airports
Marie Gilmour
Principal Consultant, Real Time Engineering Ltd
When the first recorded commercial flight took place between Boston and New York in 1927, carrying six passengers, few could have envisaged that within seventy-five years 100 million passengers in the UK alone would take to the sky.
With the burgeoning trade in charter, budget and discount airfares, airport congestion has become an increasing problem, often resulting in frustrated and dissatisfied passengers. The problem has grown with recent increases in security processes at airports. With growing emphasis placed upon customer service and satisfaction, airport management companies have come to recognise that addressing this issue is a fundamental part of the successful development and future of their industry.
Dynamic changes in passenger numbers and varying security requirements mean that airport managers have been set the continued challenge of ensuring that passenger movement through airport terminal security control areas remains smooth and uninterrupted. To keep pace with changes in the industry, capacity management techniques are in a constant state of evolution, with frequent adaptations of infrastructure and resources required to satisfy varying demand.
However, the implementation of changes, especially infrastructure changes, can disrupt the process flow and can be very costly, both in terms of expenditure and possible customer dissatisfaction. It is imperative that any changes are carried out efficiently and with minimal negative impact on the customer. This is no easy task, as anyone involved in capacity management will testify, and considering that the airport management companies have their own business targets to meet only adds to the complexity of the dilemma.
One straightforward solution to this problem would be to vary staff resources in accordance with the increase in passenger numbers, or demands on security personnel. Such measures are, however, costly to implement in terms of training and administration and offer no concrete guarantee that customer service levels will be maintained.
A simpler and more cost-effective approach would be to increase the efficiency of the resourcing function. By assessing forthcoming resource requirements, it is possible to deploy staff to locations and allocate resources where they will be most needed. There is also a requirement to react quickly to the days events which can impact passenger flow. However, in order to achieve this, operational planners need to have accurate forecasts of demand requirements.
A number of operational research techniques have been developed to assist with these forecasting activities. Most of these techniques rely on using historical data as the basis for their predictions and are used to indicate seasonal patterns in demand. These mathematical models have been refined and developed over the years, making use of up-to-date information as it becomes available. Whilst some models have been relatively successful, there is always room for improvement.
A Pragmatic Analysis
From an engineering point of view, an airport terminal shares a number of similarities with any processing facility. Both take in raw material and subject it to various checks, whereupon it is then filtered, screened and eventually results as output. This analogy, although crude, can be applied to passengers ‘flowing’ through an airport terminal. They enter the process at the point they check in, pass through various security control areas and finally exit the process when they board their intended flight. The common objective is to generate maximum output efficiency with the minimum of downtime. This should be achieved by maintaining control and consistency throughout the process flow.
In airport terms, it is imperative that lengthy queues and bottlenecks are avoided and unforeseen delays are managed effectively to keep disruption to a minimum. This is easier said than done. The dynamics of the process become very complicated when the entities being ‘processed’ exhibit individual and somewhat random characteristics. Passengers, and their behaviour, tend not to conform to any uniform standard; all have differing needs, from the seasoned business traveller, demanding rapid, efficient and professional service to the luggage-laden holidaymaker and bewildered tourist.
By its very nature, passenger behaviour will cause fluctuations in the process flow, creating peaks and troughs in activity. Add into the equation factors such as holiday periods, sporting events, seasonal anomalies or unpredictable anomalies, and the result is a constantly varying resource requirement with the knock-on effect of a fluctuating demand in the number of airport security and operational staff needed to cope.
The key to the solution is in producing accurate forecasts and, in order to achieve this, the forecasting model needs to be very flexible.
A New Solution
Real Time has been working closely with the world's leading airport operator, BAA plc, modelling passenger flows through the terminals to improve upon the current legacy systems. This forecaster, initially developed in June 2000, models passenger flows through the terminals by analysing a variety of data sources.
The innovative software solution, built on proven forecasting techniques, accurately forecasts passenger numbers at airport security areas for a given period in any given day for forthcoming seasonal flight schedules for both arriving and transferring passengers.
The project will be delivered in two phases, with an anticipated full implementation of the first phase in BAA’s south-eastern airports in May 2002.
The flexibility of the model comes from its highly configurable nature, allowing it to be adjusted to produce forecasts for specific terminal locations whilst compensating for passenger behaviour, holidays, and one-off anomalies such as security related incidents or alerts.
In addition, our knowledge and experience gained serving the transport industry has proven invaluable during the development of the forecast model. It has allowed us to take advantage of the extensive information sources currently available in BAA’s existing IT infrastructure. Information sources range from aircraft flight schedules, baggage handling and tracking systems and electronic passenger counters at controlled areas such as x-ray machines, which, when combined, offer an abundance of operational data. The combined aspects of flexibility and knowledge engineering set our model apart from competing technologies.
Early tests have indicated that the forecaster will be BAA’s most effective passenger forecasting tool to date. Bob Young, RMS development manager at BAA said, "RMS will significantly assist us in improving our customer service. It will be able to anticipate passenger demand better than existing systems and, as a result, facilitate smooth movement of passengers through the terminals."
The Complete System
The forecasting model forms part of the full Roster Management System (RMS), currently under development, comprising a number of sophisticated integrated modules built on the forecasting model. From these accurate forecasts, staff at BAA plc will be able to produce optimised staff rosters that are tailored to meet the forthcoming passenger demand.
Given an accurate forecast of passenger numbers, a conversion process is then used to translate passenger forecasts into staff and resource requirements. To achieve this, RMS will use its demand conversion and resource configuration modules. These closely related software components produce an output describing the number of staff required and how they should be best organised around the available resources.
By utilising specialised software at the heart of the roster module, RMS uses the converted passenger forecast data to automatically produce a range of staff rosters that can then be populated with staff names via the roster population module. Once produced, these rosters are not cast in stone and can easily be updated. Supervisory staff can interact with the staff deployment module, updating rosters in real-time, to suit the demands of the day. These facilities prove invaluable when there is a sudden requirement to change staffing profiles as a result of unpredictable events such as safety or security incidents.
RMS can also provide interactive administration functions providing instant feedback, allowing staff to book annual leave, request shift swaps and request overtime.
Management reports will be available on all aspects of RMS through the provision of a flexible reporting facility. This will provide an invaluable business-planning tool for BAA, furnishing them with accurate business and process information, including the facility to carry out ‘what-if’ planning scenarios.
Using Web technology, and .NET™ architecture, we are designing the system to be used in conjunction with web browser software, allowing access directly from Microsoft Internet Explorer. This will significantly reduce the commissioning effort and help minimise costs by eliminating the need for specialist client software.
The solution integrates a number of configurable modules with BAA’s existing infrastructure to provide BAA with a unique set of tools with which to manage the daily operation and deployment of staff, as well as supporting business planning by providing projected growth scenarios.
It is essential that airports address their capacity planning issues so that they can cope with the ever-increasing fluctuating passenger demand. Using advanced forecasting technology, applicable across a range of business functions, will lead to a better understanding of the business, empowering analysts with information to assist with planning and investment activities. Further more, these tools can help ensure the most effective deployment of security personnel on any day, regardless of the circumstances.
|
|
|
|