The availability of International Thermonuclear Experimental Reactor (ITER) (or an 'ITER-like') reactor will be strongly influenced by the effectiveness of the in-vessel components remote handling strategy. In the present reactor concept, the relevant key components are the divertor cassettes, located in the lower region of the Vacuum vessel. Due to erosion and damage to the divertor plasma facing components, estimated scheduled replacement of the cassettes will be required eight times during the machine lifetime. Moreover, for such an experimental tokamak where completely new plasma regimes will be established, unscheduled interventions cannot be excluded a priori. To test and optimise the divertor maintenance scenario, the so called ITER L7 R&D project has been realised at the ENEA Brash-none laboratories, with a full scale simulation of the in-vessel cassette maintenance environment and of the hot cell refurbishment operations. The basic demonstration of the validity of the current scenario has been established, and now new activities' are in progress to optimise many aspects of the operations (procedures, hardware improvements, reliability, etc.). Based on the background and the results of these activities, this paper discusses the lessons learned during the project implementation, and identifies key points of the current strategy that should be maintained for any new design of ITER or an 'ITER-like' reactor. (C) 2000 Elsevier Science B.V. All rights reserved.