1: \begin{abstract}

2: An innovative numerical technique is presented to adjust the inflow to a

3: supply chain in order to achieve a desired outflow, reducing the costs of

4: inventory, or the goods timing in warehouses.\newline

5: The supply chain is modelled by a conservation law for the density of

6: processed parts coupled to an ODE for the queue buffer occupancy. The

7: control problem is stated as the minimization of a cost functional $J$

8: measuring the queue size and the quadratic difference between the outflow

9: and the expected one. The main novelty is the extensive use of generalized

10: tangent vectors to a piecewise constant control, which represent time shifts

11: of discontinuity points.\newline

12: Such method allows convergence results and error estimates for an

13: Upwind-Euler steepest descent algorithm, which is also tested by numerical

14: simulations.

15: \keywords{Supply chains\and  ODE-PDE models\and  Optimal control\and Upwind-Euler scheme\and  Tangent vectors}

16: % \PACS{PACS code1 \and PACS code2 \and more}

17: \subclass{35L65\and 90B30\and 76N25\and 65M08}

18: 

19: \end{abstract}

20: