08.01.2020: AdONE Seminar

Prof. Teodor Gabriel Crainic (Université du Québec à Montréal): "Service Network Design and Planning Two-tier City Logistics Systems" and Alexandre Forel (TUM, AdONE): "Integrated stochastic optimization of production planning and raw material management"


Prof. Teodor Gabriel Crainic (Université du Québec à Montréal)

Service Network Design and Planning Two-tier City Logistics Systems

 Freight transportation and logistics are vital for the economy in general and cities in particular. They also yield a number of negative impacts in terms of congestion, pollution, safety, etc. City Logistics proposes new concepts and operational and business models to alleviate these negative impacts and foster an efficient and sustainable transportation system. We will review these issues, concepts and models. We will focus in particular on two-tier systems appropriate for large cities and the challenges to plan and manage such transportation systems. Service network design is a well-known methodology for planning consolidation-based transportation systems. Particular developments are required in the context of City Logistics, particularly in relation to the routing component of operations, the collaborative nature of the carriers involved, and the uncertainty in demand. We will briefly recall the service network design methodology and discuss recent results for tactical planning of two-tier City Logistics systems.

 

Alexandre Forel (TUM, AdONE)

Integrated stochastic optimization of production planning and raw material management

In the agrochemical industry, tactical production planning is done using uncertain sales forecasts as demand depends on unpredictable parameters such as meteorological conditions during the next growing season. In order to deal with the high uncertainty, forecasts are periodically updated and the production plan is implemented in a rolling-horizon fashion. However, the flexibility of the production in each month is limited by the availability of raw material. In particular, the active ingredient synthesis has long lead-times and strongly limits short-term changes. Since the active ingredient delivery depends on our production plan, the long-term decisions of today restrict the short-term flexibility of tomorrow.

Our approach includes the decisions of both production planning and raw material procurement in order to ensure the flexibility of production in rolling horizon. We formulate scenario-based stochastic models with varying level of recourse and flexibility. The models are evaluated through rolling horizon simulations and the performances measured in terms of inventory cost, customer satisfaction and stability of the production plans.

 

Date: 08.01.2020, starting at 2 p.m.

Location: City campus Z 538