Search

Most research regarding revenue management in manufacturing has considered only a short-term planning horizon, assuming supply and production data exogenously given. Motivated by the case of the Ca...

In many industries (e.g. natural resources processing, food processing, etc.), variation is intrinsic to the process. Data captured by advanced technologies and sensors can drive in-depth reflection to avoid and resolve potential issues related to process variability. In this study, we offer practical steps as a guidance for production engineers to deal with the intrinsic variability of industrial processes. We propose a methodological framework in three steps, which includes the definition and the characterization of the industrial process, predictive modeling and prescriptive analytics. In addition, we illustrate the framework application with a real case study from the textile industry.

Colloque avec actes et comité de lecture. internationale.; International audience; "This paper proposes a demand fulfillment process for Make-To-Stock environments, integrating sales and operations planning (S&OP) and order promising, for a commodity market characterized by prices and demand seasonality. Considering differentiated customers, different products and multiple sourcing locations in a multi-period context, we define a multi-level decision framework in order to support short and medium term sales decisions in a way to maximize profits and to enhance the service level offered to high-priority customers. Our research exhibits three valuable elements: (1) we developed an order promising model based on nested booking limits and which allows order reassignment i.e. changing decisions of how firm orders have to be fulfilled (2) we used a rolling horizon simulation to evaluate performance of the demand fulfillment process proposed and (3) we compare it with common fulfillment processes such first-come first-served order processing. In order to evaluate the demand fulfillment process proposed, a numerical application based on softwood lumber manufacturers located in Eastern Canada is conducted and provides evidence that better performances (overall service level, high-priority service level and overall net profit) can be achieved by using nested booking limits and reviewing previous order promising decisions whilst respecting sales commitments."

Colloque avec actes et comité de lecture. internationale.; International audience; In the lumber industry, we can observe that commodity products prices fluctuate according to seasonal patterns. Nevertheless, it is believed by the industry that it is impossible to take advantage of this information for many reasons. Firstly, the fact that many different products are produced at the same time from the same material input (coproduction) makes it difficult to produce exactly and only what is needed. Secondly, equipment is already being used at 100% capacity all year long, so there is no room to increase production when product selling price increases. Finally, the belief is that keeping finished products in stock till the moment for the right price arrives would increase inventory holding cost too much. For these reasons, the typical sawmill produces using a "push" strategy and sells its production, without much consideration of yearly price fluctuation. We have developed a mathematical model that allows planning the sales and operations of a network of sawmills at the tactical level. Using that model, we were able to show it is in fact possible to modulate production and inventory levels to increase sales revenue. We generated a single plan which, if it had been used for each of the last twelve years, would have increased the gross margin generated by an average of 1,47% of sales revenue.