Routing software in real time for Uruguay’s state oil company
At MundoGeo.com (geomatics portal) an interesting case was published about a vehicle routing system based on ESRI technology and developed with GeneXus.
A strategic advantage
By Javier Bebanz, GLP Manager and Juan Schaffner, IT Manager at DUCSA
DUCSA had decided to enter the packaged gas market but to do so, it had to enter a satisfied market and introduce something new. How did we do it? By developing a vehicle routing system based on ESRI Technology.
DUCSA is the distributing company of the Uruguayan state oil company ANCAP and its billing amounts to US$ 800 million each year. In 2004 it committed to enter the market of Liquefied Petroleum Gas (LPG) distribution which, up to that moment, was controlled by two companies that had achieved good customer satisfaction.
To this end a plan was prepared that basically consisted of two essential elements. First, to identify a strategic advantage that would be useful to compete on the market: having low cost, immediate distribution. Second, we were certain that we needed to have adequate technology.
Ingenieros Consultores Asociados (ICA), which had already worked on other implementations similar to that of DUCSA, was responsible for developing this technology that would have to enable the assignment of incoming telephone orders to the ideal vehicle in a timely manner.
Thus, in 2005 the Automated Reception and Dispatch System was created. The solution was developed with GeneXus, ArcGIS technology and fwLogistic, an ICA product, for the optimal reception and dispatch of LPG orders in real time. During peak hours, demand rises to 3,000 orders. All calls are received by an operator at DUCSA’s call center, and once the order is entered, the Optimization Engine developed with fwLogistic is responsible for informing the operator about the estimated delivery time and automatically dispatching the most adequate vehicle.
Problem facts
DUCSA has a Call Center with 24 operators, 2 dispatch stations and more than 120 vehicles of different sizes to serve more than 250,000 clients in the metropolitan area of Montevideo.
Description of the system
The objective was to build an online, real-time Automated Reception and Dispatch System. This system enables DUCSA to provide a high quality service (minimum response time, minimizing the delay from the moment the service order is entered by the user and the exact time when it is delivered by the vehicle) depending on the system workload and at low operating costs (measured in minimum total trip times).
The call is received by an operator, the call center software determines the most adequate operator and activates the system for entering LPG orders. If the telephone number detected by the call center is found on the database, the delivery address is confirmed and the order is requested, while viewing the entire history of that client. If the client is not found he/she is asked for the telephone number and other details necessary to place the order, including the delivery address which is GeoCoded by the GeoCoding Engine. Once the order is entered, it is forwarded to the optimization engine which informs the operator what the response time is.
Based on all the information entered into the system (clients, service requests, vehicles, etc.) the user responsible for planning the routes can assign service requests to the available vehicles. This planning is automatically made but it also allows for the intervention of a supervisor.
The geocoding engine is a Web Service that enables the individual geocoding of addresses as they are received. It solves both the problem of request times (reducing the number of details requested in each order) and information quality.
The Optimization engine is constantly active receiving service requests for LPG (demand) and has to determine, based on the demand and available resources, the best assignment of orders to vehicles.
It should be pointed out that future demand for services is unknown beforehand, that is, it can't be predicted when a new service request will be received (online) or from what geographic location, nor the type of service requested. At the same time, we know the exact location of the vehicles, both on duty (active) and standby (passive), through the last service address and/or departure base where it is waiting.
The best vehicle is the one that can meet the service requirement and arrives faster, driving the shortest distance possible (real distance over the city transit network).
The role of the Dispatcher (System) is to respond to the service request, that is, determine the best vehicle to meet the service request (make the assignment) and estimate how long it will take the service to reach the customer's geographic location (routing).
Based on the fact that demand is unknown, the assignment can be revoked as long as the vehicle has not reached its destination because a solution can be optimal for a given moment, but it may lose quality as time goes by depending on the data received.
The system administers the fleet of vehicles and technicians available to provide the service. Each vehicle has a stock of containers and materials for use in repairs and, in the event that they run out of stock, they must go back to the central warehouse for restocking.
When the routes are generated, the vehicle location is taken into account. This location is determined discretely according to the last customer served or a place where it was waiting.
The moment when the vehicle will be available is the estimated date and time when the vehicle will be free to take on the next job order, if it has confirmed an order or the moment when it will start the workday.
There's a System monitor that is the Geographic Information Systems interface, which enables the user to interact with the geographic elements that make up the application. It allows viewing information layers for finding locations in the city map, viewing pending service requests, viewing the location of each vehicle, making queries for information related to requests and vehicles, viewing routes made by the vehicles and planned routes.
This module provides the user in charge of planning with a powerful tool to support decision-making in order to adjust or improve the service depending on the situation at hand.
Once it has been determined which vehicle will handle the service requests, a work order is generated that must be sent to the vehicle. The administration of these associated service, dispatch, reception and event orders is carried out in a module specially designed for this purpose. Communication with vehicles is made through GSM/GPRS wireless communication technologies that make it possible to interact with the system.
Benefits
This solution has helped DUCSA to successfully enter a new market that was traditionally controlled by 2 companies which had achieved good customer satisfaction, obtaining approximately 20% of the market share.
Working with state-of-the-art technology and using a company such as ICA for the system implementation has made it possible to save costs and times in order to provide an excellent service to clients placing orders for packaged LPG with DUCSA.
Today, after using the system for two entire years, we can affirm that it has returned our investment many times over and that the decision made at the time was completely right.
By Javier Bebanz, GLP Manager and Juan Schaffner, IT Manager at DUCSA
DUCSA had decided to enter the packaged gas market but to do so, it had to enter a satisfied market and introduce something new. How did we do it? By developing a vehicle routing system based on ESRI Technology.
DUCSA is the distributing company of the Uruguayan state oil company ANCAP and its billing amounts to US$ 800 million each year. In 2004 it committed to enter the market of Liquefied Petroleum Gas (LPG) distribution which, up to that moment, was controlled by two companies that had achieved good customer satisfaction.
To this end a plan was prepared that basically consisted of two essential elements. First, to identify a strategic advantage that would be useful to compete on the market: having low cost, immediate distribution. Second, we were certain that we needed to have adequate technology.
Ingenieros Consultores Asociados (ICA), which had already worked on other implementations similar to that of DUCSA, was responsible for developing this technology that would have to enable the assignment of incoming telephone orders to the ideal vehicle in a timely manner.
Thus, in 2005 the Automated Reception and Dispatch System was created. The solution was developed with GeneXus, ArcGIS technology and fwLogistic, an ICA product, for the optimal reception and dispatch of LPG orders in real time. During peak hours, demand rises to 3,000 orders. All calls are received by an operator at DUCSA’s call center, and once the order is entered, the Optimization Engine developed with fwLogistic is responsible for informing the operator about the estimated delivery time and automatically dispatching the most adequate vehicle.
Problem facts
DUCSA has a Call Center with 24 operators, 2 dispatch stations and more than 120 vehicles of different sizes to serve more than 250,000 clients in the metropolitan area of Montevideo.
Description of the system
The objective was to build an online, real-time Automated Reception and Dispatch System. This system enables DUCSA to provide a high quality service (minimum response time, minimizing the delay from the moment the service order is entered by the user and the exact time when it is delivered by the vehicle) depending on the system workload and at low operating costs (measured in minimum total trip times).
The call is received by an operator, the call center software determines the most adequate operator and activates the system for entering LPG orders. If the telephone number detected by the call center is found on the database, the delivery address is confirmed and the order is requested, while viewing the entire history of that client. If the client is not found he/she is asked for the telephone number and other details necessary to place the order, including the delivery address which is GeoCoded by the GeoCoding Engine. Once the order is entered, it is forwarded to the optimization engine which informs the operator what the response time is.
Based on all the information entered into the system (clients, service requests, vehicles, etc.) the user responsible for planning the routes can assign service requests to the available vehicles. This planning is automatically made but it also allows for the intervention of a supervisor.
The geocoding engine is a Web Service that enables the individual geocoding of addresses as they are received. It solves both the problem of request times (reducing the number of details requested in each order) and information quality.
The Optimization engine is constantly active receiving service requests for LPG (demand) and has to determine, based on the demand and available resources, the best assignment of orders to vehicles.
It should be pointed out that future demand for services is unknown beforehand, that is, it can't be predicted when a new service request will be received (online) or from what geographic location, nor the type of service requested. At the same time, we know the exact location of the vehicles, both on duty (active) and standby (passive), through the last service address and/or departure base where it is waiting.
The best vehicle is the one that can meet the service requirement and arrives faster, driving the shortest distance possible (real distance over the city transit network).
The role of the Dispatcher (System) is to respond to the service request, that is, determine the best vehicle to meet the service request (make the assignment) and estimate how long it will take the service to reach the customer's geographic location (routing).
Based on the fact that demand is unknown, the assignment can be revoked as long as the vehicle has not reached its destination because a solution can be optimal for a given moment, but it may lose quality as time goes by depending on the data received.
The system administers the fleet of vehicles and technicians available to provide the service. Each vehicle has a stock of containers and materials for use in repairs and, in the event that they run out of stock, they must go back to the central warehouse for restocking.
When the routes are generated, the vehicle location is taken into account. This location is determined discretely according to the last customer served or a place where it was waiting.
The moment when the vehicle will be available is the estimated date and time when the vehicle will be free to take on the next job order, if it has confirmed an order or the moment when it will start the workday.
There's a System monitor that is the Geographic Information Systems interface, which enables the user to interact with the geographic elements that make up the application. It allows viewing information layers for finding locations in the city map, viewing pending service requests, viewing the location of each vehicle, making queries for information related to requests and vehicles, viewing routes made by the vehicles and planned routes.
This module provides the user in charge of planning with a powerful tool to support decision-making in order to adjust or improve the service depending on the situation at hand.
Once it has been determined which vehicle will handle the service requests, a work order is generated that must be sent to the vehicle. The administration of these associated service, dispatch, reception and event orders is carried out in a module specially designed for this purpose. Communication with vehicles is made through GSM/GPRS wireless communication technologies that make it possible to interact with the system.
Benefits
This solution has helped DUCSA to successfully enter a new market that was traditionally controlled by 2 companies which had achieved good customer satisfaction, obtaining approximately 20% of the market share.
Working with state-of-the-art technology and using a company such as ICA for the system implementation has made it possible to save costs and times in order to provide an excellent service to clients placing orders for packaged LPG with DUCSA.
Today, after using the system for two entire years, we can affirm that it has returned our investment many times over and that the decision made at the time was completely right.