The Central Logistics Project, developed by AIMMS Partner UniSoma in 2005, and is based on ABASEM, Scheduling Software developed in AIMMS for wood transportation planning. Cenibra hired UniSoma to develop and implement a Central Logistics Center to solve their wood supply logistics problem.
After harvesting Eucalyptus stands, wood is transported to the production lines at the pulp mills. Transportation can be done through roads or railways. If done by roads, this is done by a fleet of homogeneous third party road trains. In the case of railways, wood transportation is done by CVRD (Compania Vale do Rio Doce).
The regions supplied by road transportation also use intermediate courtyards. These yards are used to form strategic buffers that guarantee the regional supply whenever direct access to the wood stands is not possible. At the courtyards, intermediate or at the mills, the wood from several stands are organized in stacks. Besides theses stocks, there is wood in the fields.
To guarantee the mills demand, Cenibra uses, besides its own wood, third party wood, protected by the “custody chain”, commitment assumed by forest certification organizations (CERFLOR and FSC) that guarantee a minimal consumption of certified wood.
The trucks are loaded with the wood logs through the use of cranes. Since the wood weight is the main bottleneck for truck loading (also valid for train wagons), the harvested wood stands are submitted to a natural drying process (which lasts at least 4 weeks) before its transportation to the pulp mill.
The roadtrain’s typical cycle route is done by the following phases: travel authorization and receiving of NTM at the mill, unloading of current wood load, cleaning and refuelling of the truck, weighting at scale, traveling from the mill to the loading area, queueing for crane loading, loading, traveling back from the loading area to the mill, weighing at the scale and continuing to the chip production lines. In cases where the queue is too big, the wood load is unloading directly to the courtyard. This procedure seeks to accelerate the travel authorization of trucks for their next route and leads to an unwanted additional wood maneuvering cost.
The queueing time at the loading areas are determined by the size of the queue and the disponibility of cranes (unexpected downtime’s, operators lunch time, maintenance, etc.). The queues for the wood unloading at chip production lines define the waiting time for unloading cargo. The unloading of wood transported by trains is, preferentially, done directly in to the chip production lines.
The freight payed to third party transportation depend basically on factors that determine operational costs and machinery maintenance: roughly, the total travel distance and the access difficulties to the loading points. Besides this, there are contractual fines due to over waiting in unloading queues at mills.
Generally, the wood stand loading sequencing follows the harvesting sequence. This policy favors the minimization of freight costs (due to less moisture transportation) and quicker release of transfer areas and is only modified if the access is interrupted or when the policy conflicts with other constraints relative to the production lines supply such as the transportation capacity or the production line wood density specialization or even mix uniformity.
Cenibra hired UniSoma to develop and implement a Central Logistics Center to solve their wood supply logistics problem. This problem basically constitutes the synchronization of inflows and wood consumption at chip production lines, so as to minimize the operational supply cost, through:
- a reduction of the levels of transported wood humidity
- a reduction of fleet size (by reducing the cycle times, which basically constitutes in the reduction of queueing at loading/unloading areas)
- an increase of direct feed of production lines (less need for wood movement structure at the courtyard)
- a decrease in number of cranes due to increase in occupancy levels (also decreasing capital assets and maintenance costs)
- a reduction in crane movement in the fields
- an increase of vehicle duration due to better navigation conditions (without sharp breaking, speeding, etc.)
- Besides this, the System should provide better quality pulp, according to specialization policy or uniform mix that reduces the density variability at chip production lines
The Central Logistics Project developed by UniSoma in 2005 divides itself in to two parts: Scheduling Software (called ABASEM) and the Fleet Monitoring System (SMF).
ABASEM is a decision support tool developed in AIMMS. It contains a mathematical model based on Mixed Integer Programming techniques that generate, in an integrated way:
- the daily scheduling plan for transportation from wood stands to mills by several possible transportation methods, following a typical weekly horizon
- the stock policy for the pulp mill, stratified by wood quality, kind of forest certification ad type of wood
- daily chip production line supply scheduling
The model provides a solution that compromise between different and possibly conflicting objectives, namely.
maximization of total transported wood;
minimization of crane movement;
minimization of humidity transportation;
maximization of direct transportation of wood stands from the field to the mill;
maximization of direct feed of production lines;
minimization of wood density at production lines.
Besides this, the model respects various operation constraints such as, for example, the daily production line demand, the size of the current fleet and the custody chain.
The software was integrated with SMAD – legacy system at Cenibras forest area – for importing information regarding wood stands like current stock (in various courtyards and classified by type of wood), density, etc.
ABASEM also provides a heuristic procedure that is executed, after obtaining the daily supply plan, to automatically generate the “travel schedule”. Afterwards, this schedule is exported to the transportation companies control systems, that are in charge of executing the plan. The results obtained by ABASEM are also exported to SMAD, where a Planned X Executed control is done for activities concerning transportation.
The project also involved, besides the implantation of the Scheduling Software, the installation of on board computers BT-400 (from Blue Tec Automação Ltda.) on each of the trucks belonging to the fleet of the transportation companies, to monitor and control the steering conditions.
The steering conditions information is transmitted to a database upon the arrival of the truck at the wood receiving area, also through radio signal from the BT-400 to the base radio installed at the receiving area. More detailed information, such as motor spin values outside optimal ranges, speeding, sharp breaks and others, are communicated to the drivers through sound warnings allowing them to take immediate action, and are also registered in the BT-400.
For each region specific dry wood transportation goals were established, based on the optimal transportation schedule generated by the Scheduling Software. The result was an economy of around R$ 5 million per year (around 2.5 million dollars per year). For the transportation companies, a uniform delivery goal was established, also following the plan generated by the Scheduling Software.
Also, the following benefits were detected by the company:
- Concise view of information regarding the wood stands
- Replanning agility, when unplanned events occur
- Discrepancies concerning information available in the corporate database were exposed during the process and ended up demanding more precision in data inputs, with benefits for other company processes
- Fleet occupation information
The results from the on board computer where 2%-3% fuel economy. The steering conditions database analysis provided, for fleet managers, a powerful tool for training and allowed a change of culture for drivers, resulting in a increase of truck durability.
Walmir Coelho Lopes, Forest Operations Supervisor for Cenibra says “The project was developed during 12/2004 to 12/2005 with a broad and rich interaction between UniSoma and Cenibra teams, and is still operational up to date.”