The Transport Manager meets the following logistics objectives:

• Minimum use of transport resources such as trucks and drivers,
• Minimum impact on warehouse labour costs,
• On-time delivery of goods within specified time windows,
• Generating an accurate transport plan.

Transport Manager is a state of the art simulator based on a hybrid algorithm which generates hundreds of routes and schedules, and selects the optimal routes to satisfy user requirements of low cost and specified time constraints.

Transport Manager is initialised with a resource table and a list of loads. Resources are typically drivers and trucks based at distribution centres. Driver shift times, licence and contractual agreements can also be given as further constraints to the system. Trucks similarly have availability time windows (maintenance requirements etc.), and starting and ending locations can be specified. Loads are defined by size, volume, departure and arrival time windows for delivery and pickup; typical handling times at a depot or a delivery point are also taken into account.

Using this information the Transport Manager generates routes, and schedules all the journeys minimising the total mileage, total number of drivers and tractors used. The system merges loads, generates multidrops, backhauls and diamond journeys where appropriate, reducing empty journeys and utilising resources more efficiently. Another strength of the system is that it is flexible, and users can bias the system to reduce mileage or emphasise minimal resource usage.

The system can be used in three different operational modes. Firstly as a planning tool for estimating minimum resource requirements for a forecast period, secondly as a router/scheduler for generating a minimum cost set of journeys, and finally as a dispatcher monitoring the transport operations in real-time and intervening when necessary.

The system (Figure 1) contains the following modules:

• Intelligent Optimiser
• Graphical User Interface

The intelligent engine is a hybrid of Genetic Algorithms, Monte Carlo Simulations and Rule Based Systems. It tries out thousands of layouts and by applying the principles of evolution, it generates an optimum layout which satisfies all requirements. It uses intelligent rules in decision making to select sensible routes to reduce the computational time of the simulation.

GUI is an internet browser based module which allows local and remote access and monitoring. In its simplest form it provides spreadsheet-like tables for presenting and editing resources, time constraints and generated journeys. It can be upgraded to a map based interface if required.

Transport Manager is available in Java, C and C++, and runs on a Unix workstation or a PC. The system can easily be integrated with current databases, map based graphical user interfaces and for real-time monitoring with GPS data feeds.

The user has to provide the following:

• a database of resources and availability
• a database of list of loads and delivery time windows
• a database of locations and distances

The system is ideal for large supermarket chains and petrol stations that can meet the requirements above.

As the system uses a computationally intensive algorithm it requires high performance computers. On PCs even after a few minutes the system is able to generate a good solution. The strength of the system stems from its ability to test thousands of solutions.

Such a system will bring the following benefits:

• Minimised resource requirement,
• Reduced mileage,
• Accurate delivery times,
• Reduced environmental pollution.

We envisage three possible implementation options:

1 - Full System

This option covers the full implementation and delivery of the following modules:

a - Analysis of the current system - proposal

b - Intelligent Optimiser

c - Graphical User Interface

d - Integration with databases

e - Integration with GPS data feeds

2 - Intelligent Optimiser and Graphical User Interface

3 - Intelligent Optimiser only