D'Alembert's principle is a powerful reformulation of Newtonian dynamics that lets a moving (accelerating) system be analysed as though it were in static equilibrium. This is convenient because the well-developed tools of statics — such as summing forces and moments to zero and drawing free-body diagrams — can then be applied directly to dynamic problems.

Derivation of the required modification:

• Newton's second law for a body of mass m with acceleration a states: ΣF = m·a.
• Bringing every term to one side gives: ΣF − m·a = 0.
• The quantity −m·a is now treated as an additional force, called the inertia force (a fictitious or "reversed effective" force), equal in magnitude to m·a but pointing opposite to the acceleration.
• Including it, the equation reads ΣF + (−m·a) = 0, which has exactly the form of a static-equilibrium condition. The body is said to be in dynamic equilibrium.

So the necessary modification to Newton's second law is to incorporate a fictitious force equal to "minus mass times acceleration", i.e. the inertia force −ma.

Why the other choices are wrong:

• Omitting all resistive forces would change the physics of the problem and give incorrect results; D'Alembert's principle keeps all the real forces and merely adds the inertia term.
• Replacing all forces with their average values is not part of the principle and would generally be inaccurate for instantaneous dynamic analysis.
• Including only gravitational forces arbitrarily discards other applied and reactive forces and is not what the principle prescribes.

Therefore the correct modification is to add the inertia force of magnitude −ma.