In most simple accept or reject decisions, IRR and NPV will select the same project. However, NPV has certain advantages over IRR.

The NPV of a proposed project, if calculated at an appropriate cost of capital, is equal to the increase in shareholder wealth which the project offers. In this way, NPV is directly linked to the assumed objective of the company, the maximisation of shareholder wealth.IRR calculates the rate of return on projects, and although this can show the attractiveness of the project to shareholders, it does not
measure the absolute increase in wealth which the project offers.
NPV looks at absolute increases in wealth and thus can be used to compare projects of different sizes. IRR looks at relative rates of return and in doing so ignores the size of the investment projects.
NPV is not subject to the technical difficulties which limit the usefulness of the IRR method.

First, in situations involving multiple reversals in project cash flows, it is possible that the IRR method may produce multiple IRRs (that is, there can be more than one interest rate which would produce an NPV of zero). If decision makers are aware of the existence of multiple IRRs, it is still possible for them to make the correct accept or reject decision using IRR, but if unaware, they could make the wrong decision.

Second, in situations of mutually exclusive projects, it is possible that the IRR method will (incorrectly) rank projects in a different order to the NPV method. This is due to the inbuilt reinvestment assumption of the IRR method. The IRR method assumes that any net cash inflows generated during the life of the project will be reinvested at the project’s IRR. NPV, on the other hand, assumes a reinvestment rate equal to the cost of capital. Generally NPV’s assumed reinvestment rate is more realistic and hence it ranks projects correctly.

Finally, NPV can be used in situations where the cost of capital changes from year to year. Although IRR can be calculated in these circumstances, it can be difficult to make accept or reject decisions as it is difficult to know which cost of capital to compare it with.