Final answer:
In linear programming, the non-negativity constraint disallows negative variables, making option 1 not acceptable. Upper bound and equality constraints are valid.
Step-by-step explanation:
Among the options provided, the one not acceptable as a constraint in a linear programming problem, specifically a minimization problem, is a negative variable. In linear programming, all variables are generally required to be non-negative, meaning they cannot take on negative values. This fundamental rule is known as the non-negativity constraint. An upper bound constraint, which limits the maximum value a variable can take, and an equality constraint, which specifies that the value of a variable must be equal to a certain number, are both acceptable in linear programming.