The overview efficiency and effectiveness of an entire organization depends greatly on how well its
resources are being utilized. Resources may include personnel, equipment, vehicles, space,
warehouses, etc. Sufficient resources must be deployed to meet business demands as well as ensure
all business, marketing and operational rules and constraints are satisfied.
Our AI Resource Allocation Engine makes use of a variety of advanced AI techniques, such as
constraint programming, business rules, and genetic algorithms to intelligiently produce optimized
resource allocation schedules and plans that precisely match target demands while satisfying all
rules and constraints. Our AI technology also allows businesses to react to changes in workload
demands or unexpected events through an intelligent dynamic rescheduling algorithm.
The following are some success stories of deployed AI resource allocation systems using our
Hong Kong MTR
Possession/Engineering Work Scheduling (2010)
In 2007, the KCRC merged with MTR to oversee all the railway lines in Hong Kong. After the merger,
MTR operates 10 heavy rail lines and 12 light rail lines, spanning 218.2km and 152 stations. In
2010, MTR decided to create one AI system to schedule and manage all engineering works for all the
rail lines. This new system is an enhanced and updated version of the AI Engine deployed in 2004
for the subway system.
The new enhanced AI Engine has been in daily use to schedule and manage all engineering works for
all the rail lines in Hong Kong since July 2013. All engineering works scheduled are ensured to be
safe and satisfying all business and operational constraints. The schedule is also optimized so
that MTR can perform more engineering works with its existing resources. In addition, valuable
domain knowledge and expertise related to these regulations and guidelines are now quantified,
coded and preserved within the organization, for use by anyone using the system.
Hong Kong MTR
Possession/Engineering Work Scheduling (2004)
MTR carries 2.4 million passengers daily, making it one of the busiest in the world. Despite the
large traffic-flow, the MTR has set for itself a very high service quality standard - train
punctuality and delivery must be 99% and 99.5% on time respectively. To ensure these standards are
met, the scheduling of engineering works must be optimized so that all necessary maintenance tasks
are done in a timely manner.
Through the use of AI techniques, we were able to help the subway streamline their
scheduling/rescheduling processes and maximize their resource utilization, while providing early
identification of potential violations of safety and operational regulations and guidelines for all
scheduled engineering works. In addition, valuable domain knowledge and expertise related to these
regulations and guidelines are now quantified, coded and preserved within the organization, for use
by this and other systems.
Hong Kong Airport Authority
Airport Bay/Stand Allocation
The Hong Kong International Airport has facilities to handle over 35 million passengers and 3
million tones of air cargo annually. Efficient use of these resources is crucial and critical to
achieving the required throughput and service commitments of HKIA. The most important resource at
the airport is, of course, the aircraft parking stands.
Aircraft stands are allocated daily based on the flight schedule and a set of intricate operational
constraints. These constraints ensure airport safety and passenger convenience, and facilitate
smooth operations for airlines and handling agents.
Our Stand Allocation System (SAS) is a mission critical system designed for non-stop 24-hour daily
Our AI Scheduling Engine consists for custom designed scheduling/rescheduling algorithms using AI
constraint programming technology.
Hong Kong Civil Aviation Department
Airport Check-in Counter/Desk Allocation
In 1994, CAD had to turn down thousands of flight requests per year due to limited airport
resources. The aging airport had a small terminal building that was already over congested.
Although air traffic was on the increase, facilities in the passenger terminal building was fixed
and limited. In order to service more flights and more passengers, resources such as check-in
counters must be allocated very efficiently.
The daily assignments of check-in counters to airlines and handling agents were the responsibility
of CAD. This assignment is based on the daily flight schedule, usage patterns and statistics, and a
set of operational and business rules, constraints and parameters.
The problem complexity made it very difficult for a human to produce an optimal plan. Moreover, the
manual process was too time-consuming to be effective. In order to cope with an increasing amount
of passengers using the airport, CAD decided to implement an automated solution in late 1994.
The Computerized Check-in Counter Allocation System (CCCAS) maximizes the utilization of airport
facilities by optimizing the location and number of check-in counters assigned to each flight. We
uniquely combined computer simulation with advanced constraint-based optimization techniques,
provided by our AI Scheduling Engine, to produce a highly efficient scheduling algorithm.
Hong Kong Air Cargo Terminals Ltd
Air Cargo Handling System
In 1995, HACTL started to investigate how the next generation resource allocation system should be
designed and built to support its huge new air cargo handling facility, Super Terminal 1 at the new
Chek Lap Kok International Airport.
The new SuperTerminal 1 was designed to handle 2.6 million tones of air cargo
yearly, which will meet the business needs in Hong Kong and Southern China well into the next
decade. Therefore HACTL needed effective scheduling methodologies to support its new Super Terminal
Our scheduling methodology based on advanced constraint-programming and object technology. A
prototype system was developed to demonstrate the feasibility of the methodology using a
scaled-down version of the new Super Terminal 1. This prototype assigns resources such as conveyor
belts, hoists, cranes, and decks to service containers of import and export flights. The system
performs scheduling based upon expected number of containers and the resources required by each
flight. This scheduling system also performs real-time reactive scheduling to accommodate changes
in allotment and resource availability.
We also applied the proposed scheduling methodology to produce a design for a workspace and
personnel scheduling system. The systems allocate suitable work areas for the build-up or the
break-up of each container and a team of workers to perform these tasks based on operational
constraints and criteria.
Modern Terminals Ltd
Container Terminal Berth Allocation
MTL owns and operates five main berths and one feeder berth where up to eight vessels can berth
simultaneously. With rapid business growth, these berth spaces must be allocated very efficiently
to maximize utilization.
Striving for performance is not easy. In 1996, MTL commissioned us to design and develop an
Artificial Intelligence (AI) pilot system to automatically assign and allocate berth space based on
demand, usage patterns, and operational and business rules and constraints.
Combining advanced scheduling technology with constraint programming, we designed a multi-user
application that captures all the rules and constraints related to berth allocation and its
management. Our system produced allocation plans in seconds, compared with hours that are normally
required if done manually.