by Renaud De Landtsheer, Christophe Ponsard and Yoann Guyot, CETIC

The efficient operation of logistics and supply chain systems requires businesses to solve large and complex optimisation problems such as production planning, fleet management and order picking. The OscaR Open Source framework supports the optimisation of such problems through a variety of powerful solvers offering various expressivity, optimality, and scalability trade-offs.

Optimisation problems often arise in logistics and supply chain systems. Although a number of “off-the-shelf” commercial solutions exist, companies may have problems using them for various reasons: their problem might be too specific to be managed within a closed solution, or its cost might be too high for smaller structures. The multiplicity of problems might also require the use of different techniques or even a hybrid approach based on a combination of methods, which is also difficult to achieve in a closed framework.

OscaR is an Open Source software framework [L1] developed jointly by the CETIC research centre, the University of Louvain and the N-SIDE company. CETIC is contributing through the SimQRi (Simulative Quantification of procurement induced Risk consequences and treatment impact in complex process chains) [L2] and TANGO (Transparent heterogeneous hardware Architecture deployment for eNergy Gain in Operation) projects [L3] . The framework notably supports constraint programming, constraint-based local search, mixed integer programming and discrete event simulation, and is especially tailored for combinatorial optimisation. Here we illustrate the use of two relevant OscaR modules on specific case studies in logistics.

Managing procurement risks through Discrete Event Simulation (DES)
Decision makers in the field of procurement and production logistics need to design cost optimal processes taking into consideration the firm’s exposure to supply risks. Assessing risks in logistics processes requires quantification of the impacts at the delivery side (e.g. order delay, quality, quantity) of several factors, preferably in a statistical way. This requires the efficient simulation and computation of key indicators on the global logistics system. To address these issues, OscaR.DES is able to capture the problem directly at the logistics domain using both textual and graphical primitives (see Figure 1) such as suppliers, storage, production processes, order policies, etc. Based on this, the DES engine can very efficiently compute the system evolution by evaluating changes only when they occur and updating complex risk indicators combining basic properties (process delays, storages volume…) using a rich set of operations (logical, arithmetic, temporal,…). Large sets of histories can then be aggregated through Monte-Carlo techniques, e.g. to explore the procurement policies that are able to best address the identified risks. Figure 1 illustrates a typical risk-oriented model for a complex assembly process.

Figure 1: Supply chain model for a complex assembly.
Figure 1: Supply chain model for a complex assembly.

Optimising vehicle fleet routing
Local search techniques are known to be efficient at solving such problems - especially large problems. The CBLS engine of OscaR supports the modular specification of routing problems by assembling problem elements (such as objective functions, strong and weak constraints, time windows, traffic jams, SLA as shown in Figure 2). On the other hand, it provides a rich domain specific language for working out powerful search procedures combining a rich set of neighbourhoods (e.g. insertPoint, onePointMove, threeOpt) in a declarative style using movement, acceptor and meta-heuristic operators (e.g. tabu search, simulated annealing). This not only results in very efficient search procedures expressed at the logistics domain level but also reduces development time and facilitates maintenance when a problem evolves.

Figure 2: Library of reusable VRP problem elements.
Figure 2: Library of reusable VRP problem elements.


[1] L. M. P. Van Hentenryck: “Constraint-based Local Search”, MIT Press, 2009.
[2] R. De Landtsheer et al.: “A Discrete Event Simulation Approach for Quantifying Risks in Manufacturing Processes”, Int. Conf. on Operations Research and Enterprise Systems, February 2016.
[3] R. De Landtsheer et al.: “Combining Neighborhoods into Local Search Strategies”, 11th MetaHeuristics International Conference, June 2015.

Please contact:
Renaud De Landtsheer
CETIC, Belgium
Tel: +32 472 56 90 99
This email address is being protected from spambots. You need JavaScript enabled to view it.

Next issue: January 2018
Special theme:
Quantum Computing
Call for the next issue
Image ERCIM News 105 epub
This issue in ePub format

Get the latest issue to your desktop
RSS Feed