Antimicrobial resistance (AMR) is a substantial global One Health problem. This paper reports on initial, proof-of-concept development of an agent-based model (ABM) as part of wider modelling efforts to support collaborations between groups interested in policy development for animal health and food systems. The model simulates AMR in poultry production in Senegal. It simultaneously addresses current policy issues, builds on existing modelling in the domain and describes AMR in the broiler chicken production cycle as seen by producers and veterinarians. This enables implementation and assessment of producer antimicrobial use (AMU) and infection prevention and control (IPC) strategies in terms of immediate economic incentives, potentially helping to advance conversations by addressing national policy priorities. Our model is presented as a flexible tool with promise for extension as part of AMR policy development in Senegal and West Africa, using participatory approaches. This work indicates that ABM can potentially play a useful role in fostering counter-AMR initiatives driven by food system actor behaviour in lower- and middle-income countries (LMIC) more generally.

In this article, the problems studied are to optimize the life of the network of sensors covering mobile targets by minimizing the energy consumption of the network. Indeed, wireless sensor networks have received particular attention in recent years, as their applications are common today, such as vehicle tracking or battlefield monitoring. In a wireless sensor network, the sensor battery plays an important role. The function of the sensor depends on the battery life. Replacement of batteries becomes impossible once deployed in a remote or unattended location. A set of randomly placed sensors monitors targets moving in an area. Each sensor has a limited lifetime and two states: Active or Inactive. An active sensor can monitor targets in its monitoring sub-area at a necessary energy consumption cost. To model these problems, we use a mathematical model based on linear programming with mixed integer variables. AMPL is used to formulate the model and MINOS for its resolution and the numerical results allowed to obtain an adequate activation scheduling of the sensors by improving their lifetime.