No abstract

Continuous timely repair and replacement of infrastructures, equipment and utilities play an important role in maintaining the smooth-running of a city or local community. Thereby, to help individuals and businesses go about their daily activities with ease, it is vital to develop a proper method for automatically identifying and assigning capable workers for tasks. This paper defines the community management service task allocation problem as CMS-TAP and hence an end-to-end “recommendation + allocation” network, i.e. a task recommender and allocation optimization network (denoted as TROpt-NET), is then developed for handling such problem. TROpt-NET consists of two layers, namely one for predicting worker ability and the other for allocating tasks which are TR Layer and TA Layer, corresponding to “recommendation” and “allocation” of tasks. Different from operations research approaches where workers are assigned to jobs based on their pre-labelled skills and fixed locations, we propose a task recommender and allocation optimization network. The TR layer is a task recommender system designed to learn implicit worker abilities for different tasks using Neural Collaborative Filtering (NCF) by mining a historical dataset of worker task completion. Whereas in the TA layer a differential optimization approach for allocation is used because of its differentiable property and ability to allow for backpropagation to the prediction layer. In this study, we first formulate the CMS-TAP problem as a recommendation +optimization problem and then propose and end-to-end network architecture that tackles the problem in a real-world setting. TROpt-NET curbs uncertainty and assumptions in optimization by learning to more accurately approximate worker ability across different tasks. Additionally, the network can learn implicit worker abilities enabling optimal utilization of workers across a wide range of tasks, which is often ignored in task allocation problems. We find that normalizing worker ability across all tasks improves the implicit learning capability of the network and that good approximations don’t always lead to optimal allocation but learning allocations by backpropagating through recommendations improves the allocation objective. Offline experiments on a real-world large-scale dataset demonstrate the effectiveness of our proposed TROpt-NET.

One of the main concerns with Particle Swarm Optimization (PSO) is to increase or maintain diversity during search in order to avoid premature convergence. In this study, a Performance Class-Based learning PSO (PCB-PSO) algorithm is proposed, that not only increases and maintains swarm diversity but also improves exploration and exploitation while speeding up convergence simultaneously. In the PCB-PSO algorithm, each particle belongs to a class based on its fitness value and particles might change classes at evolutionary stages or search step based on their updated position. The particles are divided into an upper, middle and lower. In the upper class are particles with top fitness values, the middle are those with average while particles in the bottom class are the worst performing in the swarm. The number of particles in each group is predetermined. Each class has a unique learning strategy designed specifically for a given task. The upper class is designed to converge towards the best solution found, Middle class particles exploit the search space while lower class particles explore. The algorithm’s strength is its flexibility and robustness as the population of each class allows us to prioritize a desired swarm behavior. The Algorithm is tested on a set of 8 benchmark functions which have generally proven to be difficult to optimize. The algorithm is able to be on par with some cutting edge PSO variants and outperforms other swarm and evolutionary algorithms on a number of functions. On complex multimodal functions, it is able to outperform other PSO variants showing its ability to escape local optima solutions.

No abstract