In the framework of the EUROfusion Programme, EU is preparing the conceptual design of the inner fuel cycle of a pulsed tokamak DEMO. This paper illustrates a quantified process to shape a R&D programme that exploits as much as possible previous R&D. In an initial step, the high-level requirements are collected and a novel DEMO inner fuel cycle architecture with its three subsystems vacuum pumping, matter injection (fuelling and injection of plasma enhancement gases) and tritium systems (tritium plant and breeder coolant purification) is delineated, driven by the DEMO key challenge to reduce tritium inventory. Then, a technology survey is carried out to review potential existing solutions for the required process functions and to assess their maturity and risks. Finally, a decision-making scheme is applied to select the most promising candidates. ITER technology is exploited where possible. As a primary result, a fuel cycle architecture is suggested with an advanced tritium plant that avoids full isotope separation in the main loop and with a Direct Internal Recycling path in the vacuum systems to shorten cycle times. For core fuelling, classical inboard pellet injection technology is selected, in principle similar to that proposed for ITER but aiming for higher launch speeds to achieve deep fuelling of the DEMO plasma. Based on these findings, a tailored R&D programme is shaped that tackles the key questions until 2020.