Multi-Objective Task Allocation for Dynamic IoT Networks

Survey on Multi-Objective Task Allocation Algorithms for IoT Networks

“…Network Lifetime is defined as the difference between the latest end time of the last valid allocation and the start time of the first valid allocation [ 9 ]: …”

“…The latency of the allocation is defined as the maximum latency of all connected tasks. The Latency of an allocation series is the maximum latency of any allocation of the series [ 9 ]: …”

“…Availability is defined based on the error indicator function e , see Equation ( 6 ), as the fraction of time when the network is fulfilling the task T (i.e., the network is error-free ), divided by Network Lifetime ( ) [ 9 ]: …”

“…Weikert et al [ 9 , 11 , 24 , 25 ] proposed a group of EAs suited to deal with the challenges of task allocation in mobile and failure-prone networks. Their latest work [ 9 ] combines these approaches and improves upon the weaknesses of the earlier versions. Their work also incorporates a true MOO approach, optimizing network lifetime, latency and availability.…”

“…Out of the 37 reviewed works, only 4 are fully suitable for networks that include node failures. This is either achieved through retaining a large archive of previously optimized solutions [ 9 , 25 ], which has the drawbacks of increased computational effort to keep an optimized diverse archive as well as not guaranteeing an suitable solution in the archive, especially if there is a high rate of failure in the network. Introducing redundancy in the task allocations [ 26 , 39 ] is another option used to mitigate node failures, but comes at the cost of higher energy consumption and a high rate of failures can still cause task outages if all assigned nodes fail.…”