5.2. Big data in DT DT is used in sectors which have multiple components resulting in multiple parameters. Hence the data collected from these sources ends up being a large high-dimensional dataset. Moreover, if the time frequencies of the data collected from these different components do not match, the resulting data can be fragmented. Therefore, there exist time lags in time series data. Additionally, collection of data from multiple inter-connected and not-connected components, with highstream synchronisation and integrating this data, is a challenging task in terms of technology, implementation, cost and resources . 6. Challenges in DT Min et al. identify two major issues related to implementing Currently DT models face the following challenges, some of which are weighed more depending on the domain the DT is being implemented. These challenges are majorly technical:

As a dynamic environment requires a well-researched tool, better concrete and practical frameworks are needed for big data application to the continuously changing environment of DT. 2. Data processing issues for time series data: Data gathered from IoT devices in the factory have large dimensions. Moreover, the data collected may have different time cycles. 1. High-fidelity 2-way synchronisation is especially hard for largescale industries, requires resources and high-stream IoT connection [1,35,47]. 2. Interoperability with existing software being used in a production lifecyle : Industries use various software for tasks such as inventory, product management, operations. The compatibility of DT with these is a challenging issue, tackling which might lead to delay in implementations.

The authors identify the above problems and solve the problem of using data from different time frequencies, by proposing a method to generate same frequency time series data.

3. Cybersecurity concerns, IoT security, cross industrial partners security : With the digital twin operating across multiple industrial partners and inventory sites, the security concerns are inevitable. Not only the cross industry security concerns but also the leak of real-time monitoring data can be hazardous to a firm. 4. Add-Ons: Using DT entails certain add-ons like cost, resources and research. Since implementing DT and profiting from it is a timely process, DT can be costly if the life and span of a project are short. Building a software for DT also demands a team of programmers, developers and domain experts to test the suitability of the software for the particular task. Moreover, like any technology, DT also needs to be updated according to the recent developments in the technologies it relies on (IoT, big data, machine learning). Industries with long-term DT use will therefore need to continuously invest in this research, which might lead to added cost. As DT requ