CONTEXTIn northern West Bengal and southern Bangladesh, Rabi crops, sown in winter and harvested in the spring, are an important source of income and nutrition for the target communities. In the study areas, NGOs and extension services have been engaging with farming communities on selecting suitable crops for the upcoming season. This engagement took place in the absence of quantitative tools to discuss trade offs and what-if scenarios to support an informed discussion. OBJECTIVEThe objective of this study was to design a crop-choice model to support extension agronomists in engaging with the farming communities more effectively using a quantitative analysis tool. In this process, we explored how agricultural system models can be inclusive and allow participation and eventual application by NGO and government extension agents, using a process of gender sensitive contribution. MethodsThe crop-choice model includes several constraints related to available resources, including land, labour, capital and technologies to calculate optimal crop areas within the Rabi season. The crop choices are determined considering gender specific community perceptions of risks, labour use, market, price-volatility, self-consumption, water use and cash flow. The model was used to explore the consequences of different crop choices on income, gender specific labour, use of inputs and markets, and to reveal the trade-offs of pursuing different crop choice pathways in the context of agricultural intensification. The initial model was demonstrated to the farmers and the extension agents, and based on their reflections, it was fine-tuned further to make the engagement process more effective. For better communication with the participating farmers, the model related questions and findings were translated in the local language (Bengali). RESULTS AND CONCLUSIONSWe found that the modelling process can be made more inclusive from the outset by including farmers, NGO and extension agents as co-contributors of the model at each of the modelling steps and incorporating their reflections. Such an inclusive and reflective approach provides easy-to-use interfaces and enables translation of model results in ways that more effectively benefit farming communities. The process of engagement with farmers during tool development has been very valuable to farmers and researchers alike and the use of the tool has made the farmers informed about alternative scenarios and led to actual benefits for crop choice decision making. Inclusive farming systems models need to consider the gender dimension and its critical role in farm decision making and how this can be included in models to reflect the diversity of decision process. To maximise relevance of the model for next users and farmers, it was important that these key stakeholders were part of the model development from the beginning, this study has focused on the same throughout. SIGNIFICANCEThis study demonstrated that appropriate design and development principles enabled bioeconomic farming systems models to be used by NGO and government extension agents to engage with farming communities as discussion support tools in farming decisions. Such an approach would make the engagement process more convincing and effective. Besides, it would also provide useful insights to the NGO and extension agents in revising their intervention strategies.