*2.2. Literature Review*

Data envelopment analysis (DEA) is a linear programming method for measuring the performance of multiple industry players (DMUs) when a production process presents a structure of multiple inputs and multiple outputs. DEA has been widely applied by many scientists in many different scientific disciplines [5–8]. Chowdhury et al., (2011) used the Malmquist productivity index to evaluate the performance of hospitals in Ontario between 2003–2006 [9]. In that study, the authors did an in-depth analysis of the efficiency of technical and technological investment in hospitals in Ontario. The researchers pointed out the limitations of technical and technology investment in these hospitals, and proposed solutions to improve the efficiency of technology investment. Abbas et al., (2015) used the Malmquist productivity index to evaluate and compare the performance of Islamic banks and conventional commercial banks [10]. The results show that Islamic banks have higher operational efficiency than conventional commercial banks. The reason is that, in Muslim countries, Islamic banks are sponsored.

Bahrini et al., (2015) used the Malmquist productivity index to evaluate the performance of 33 Islamic banks in 10 countries around the world for the period 2006–2011 [11]. The results show that Islamic banks were hit hard by the 2008 global financial crisis. In that study, the researchers also proposed solutions to help Islamic commercial banks' recovery and growth. Lee et al., (2017) used DEA models to evaluate the performance of 18 banks in Korea in three categories: National Bank, Regional Bank, and Special Bank [12]. The authors found that special banks have the highest business efficiency compared to other types of banks. Wang et al., (2017) used DEA models to evaluate the performance of infrastructure investment and development companies in Vietnam [13]. The authors evaluated and selected good investors and suggested that the Vietnamese government establish appropriate policies when selecting contractors for infrastructure investment projects to achieve good results on time and on budget.

Grey system theory was introduced by Professor Julong Deng in 1982 [14]. Grey prediction theory is a multidisciplinary forecasting science that has been applied in almost all sciences. Since its introduction, it has been applied by scientific works around the world [15–19]. Fan et al., (2018) used a Grey forecasting model to forecast natural gas demand in China [20]. The results of that research helped the Chinese government to develop and issue energy policies to ensure a stable supply of natural gas for production and daily life. Nguyen (2020) used the super-slack-based model to select partners for construction companies in Vietnam [21]. After selecting an alliance partner, the author used two forecasting methods, the Grey forecasting model and ARIMA model, to forecast the business situation of construction companies. Wang et al., (2021) used the Grey forecasting model to predict the number of railway passengers by quarter (period 2020–2022) in China. This result helps railway management companies create plans to meet the travel needs of customers.

In this study, the Malmquist productivity index (MPI) is used to evaluate the business performance of logistics enterprises in Vietnam from 2017 to 2020. The authors continue to use the super-slack-based model to select optimal partners to enter strategic alliances with businesses to help businesses recover and develop sustainably after the pandemic. In addition, in this study, the authors used the Grey forecasting model simultaneously to forecast and provide a post-pandemic picture of businesses participating in the alliance, giving managers a solid basis when making decisions about implementing alliances.
