“…From a mathematical point of view, our model is identical to the acyclic permission structures introduced in Gilles et al (1992) and the structures modeling precedence constraints introduced in Faigle and Kern (1992), even if these structures are sometimes defined by means of digraphs. These two models have been extensively studied in the literature (van den Brink, 2017;Algaba et al, 2017, survey some results) and have in common that the associated structure imposes some restriction on the formation of coalitions or on the subset of coalitions to which an allocation rule is sensitive. In van den Brink and Gilles (1996), van den Brink (1997) and van den Brink and Dietz (2014), the so-called conjunctive, disjunctive and local permission values are computed as the Shapley value of a restricted game in which only the "feasible" part of a coalition is productive, where the "feasible" part is the largest subset of the coalition that contains all the hierarchical superiors deemed necessary for the worth generation (these vary depending on the model: conjunctive, disjunctive or local).…”