Order picking operations have attracted a lot of attention from researchers in the last decades as these are identified as the most time-consuming, labour-intensive and expensive activities for most warehouses. To remain competitive, warehouse managers always keep searching for new alternatives of picking systems with the goal of increasing picking productivity without raising significantly their expenses. However, investing in a new picking system requires gaining insight from the different options beforehand, which is difficult without testing the latter. Decision-makers are therefore skeptical about whether the gain in productivity is worth the related investments.
In this recent era of Industry 4.0, numerous innovative technologies have emerged, with augmented reality (AR) being one of the most promising ones as it can be applicable in many fields. Paired with AR, smart glasses present a great potential to assist operators in warehouse picking activities as they do not require the use of hands while having the abilities to display relevant information to the picking locations at eye-level.
The present paper is then focused on the use of smart glasses in order picking operations. The main objective of the study is to perform an assessment of a pick-by-vision system from both productivity and economic perspectives and to compare it with other picking systems. The inclusion of the economic perspective in the evaluation of the pick-by-vision system is the main contribution of this research.
The methodology has been the following. First, the literature has been the main source of input to know the state-of-the-art of pick-by-vision systems. A pick-by-vision system using the Hololens was then developed for this specific study in order to conduct the evaluation. To measure its picking performance in terms of picking time and picking errors, several test people were solicited to accomplish order picking tasks using the developed Hololens solution in the pilot warehouse set up in the Logistics 4.0 laboratory. Following the experiment, results from the measurements were integrated in an economic model which calculates an hourly cost function of a given picking system. A comparative analysis was then conducted with five other picking systems which are barcode handheld, RFID tags handheld, pick-by-voice, pick-by-light and RFID pick-by-light systems, built on the work of some other researchers.
From the calculations of the hourly cost functions of each picking system, it has been possible to determine the most convenient picking system depending on the demand from the customer orders (translated in the number of requested picking rows). The considered pick-by-vision revealed to be the most profitable system for most of the cases where the number of requested row n_R was lower than 170 rows/hour, but was outperformed by the RFID pick-by-light for higher n_Rs. A qualitative assessment of the pick-by-vision system was also added in the discussion. These findings can be beneficial to warehouse managers who are in the process of deciding for investments in a new picking system.