Digital Planning Support for Modular Logistics Plants in the Process Industry

2.1 Modular Production

Conventional production systems in the process industry only meet the market complexity drivers described in Section 1 to a limited extent [^{10, 14}]. Even though continuously operated single-product plants are very efficient, they are also largely inflexible in terms of output, product range, and relocation, whereas discontinuously operated multi-product plants offer flexibility with regard to the product range within certain operating windows but are not as efficient [^{10, 14}]. The focus of industrial and research interests is therefore to combine the advantages of both approaches, which means modular, adaptable, and decentralized plant concepts with small to medium capacity profiles, leading to greater flexibility [^{5, 9}].

Modularization in this context means the process of dividing a production system into modules, which provide one or more specific functions [^{15, 16}]. The vision of the modular approach is that the individual modules can be exchanged or moved quickly and easily with existing standardized interfaces according to the “Plug & Operate” principle [^{9, 17}]. This enables the creation of completely new production systems without the need for a new validation of the entire production process [^{2, 18}]. Using modules enables a faster market entry time [⁹]. Plant capacity can be flexibly adjusted to demand by numbering modules up or down, allowing for gradual and market-oriented scaling, enabling temporal and spatial market tracking [^{7, 11}].

In modularization, a distinction is made between physical modularization and automation modularization. Physical modularization focuses on modularizing the process engineering equipment and apparatuses into separate, encapsulated units, known as modules [⁹]. According to VDI2776, experts divide modular production plants into a basic structure consisting of modular plant (MP), process equipment assembly (PEA), functional equipment assembly (FEA), and components [⁶]. In the course of physical modularization, container solutions are being discussed in which—depending on the application—PEAs, FEAs, and components are located and together form an MP [⁹].

In order to automate production using modular “Plug & Operate” principles, the process chain also needs to be automated in a modularized way [^{3, 19}]. However, conventional process control systems do not meet the requirements of modular production plants as outlined in the VDI/VDE/NAMUR 2658, a development guideline for successfully applying MPs, particularly in establishing a common interface standard [²⁰]. A key enabler is a service-oriented architecture, where each module—also referred to as PEA—provides encapsulated process functions as services, such as starting, stopping, pausing, or holding [^{21, 22}]. These modules are described in a functional, vendor-independent module description, the so-called module-type package (MTP), and integrated into the so-called process orchestration layer (POL), a higher level administration, coordination, and monitoring layer [^{3, 23}]. Although the MTP thus enables the seamless integration of modules into the POL [²²], the POL configures a specific MP topology in order to execute the module services provided by the PEAs in the sequence required for the production process [^{24, 25}].

Various demonstrators and pilot projects have already proven the potential of modular production in the process industry. In research, the projects F³Factory, CoPIRIDE, and the ENPRO initiative are worth mentioning [^{7, 9}]. Major players in the process industry, including BASF, Bayer, Evonik, Merck, Procter & Gamble, and AstraZeneca, have implemented prototypical modular production concepts [^{2, 3, 21}].

2.2 Modular Logistics

Conventional logistics plants in the process industry offer low flexibility [²⁶]. To meet volatile system requirements, such as variable demand quantities, product, or packaging types [^{5, 14}], extensive technical changes are required [²⁶]. Setup, cleaning, and maintenance measures must also be planned with sufficient lead time and lead to long downtimes [²⁶] and underutilized capacities [²⁷]. Because modular production can only reach its full potential if logistics plants have the same degree of modularization, research increasingly focuses on modular intralogistics solutions as they promise high transformability under volatile system requirements [²⁶].

However, flexible, modular solutions only exist for limited areas of intralogistics systems. Researchers and developers have already conceptually created container-based logistics modules for provisioning, conveying, and storage as well as cleaning, filling, and packaging [^{11, 16}]. As a result, Pannok et al. present a procedure for designing flexible intralogistics units for the supply and disposal of modular production systems, supporting operators in aligning their intralogistics processes with system requirements [²⁸]. Gryczycha et al. have developed a reference process model for the modularization of production-related logistics through process standardization [²⁹]. So far, developments focus on a flexible design of single modules [^{5, 11}]. To ensure fast and compatible (re-)configurability in dynamic environments and to reduce planning and commissioning efforts, these systems need modular-enabled automation solutions with “Plug & Operate” capabilities [^{5, 11}]. The following will outline the most important technical requirements of modular logistics.

The “Plug & Operate” capability of logistics modules requires compatibility through standardized automation interfaces, ensuring seamless integration on both horizontal (adjacent production and logistics modules) and vertical (control system) levels. To minimize engineering effort, integration must be simplified and automated as much as possible. The MTP concept described in Section 2.1 is also fundamentally suitable for the automation of modular production-related logistics plants [^{30, 31}]. The logistics system adopts the basic structure of MPs by transferring so-called Logistics Equipment Assemblies, which offer their process functions as services [³¹]. The NAMUR recommendation NE171 outlines the automation requirements for this approach [³²]. Blumenstein et al. outline the functional scope of a Logistics Orchestration Layer as a counterpart to the POL [²³]. A first demonstrator has implemented this concept, focusing on rapid reconfigurability through modular automation [²⁷]. Increasing automation flexibility enables higher mobility and flexibility by reducing assembly, dismantling, and set-up times [²⁶]. However, mechanical and electrotechnical integration remains a challenge, and modular safety concepts must be developed to ensure machine safety.

2.3 Existing Planning Approaches in the Process Industry

Conventional planning methods rely on tailor-made process design, i.e., equipment and systems are designed for an exact operating point [⁹]. In contrast, MP planning adapts process parameters to standardized equipment [⁹], reusing cross-project planning knowledge and increasing workflow parallelization [³³]. This approach enables early knowledge of available apparatuses, reducing planning time but potentially compromising process efficiency [⁹]. Conventional planning methods are unsuitable for shortening planning times or reducing investment costs through standardization [³³]. Ultimately, it can be stated that the modular design of a plant has a significant impact on the planning process [⁹].

In module-based planning, the focus is on reuse of planning content and reduced times in process design resulting from the use of pre-engineered equipment [³⁴]. The engineering of MPs consists of project-independent module-type engineering and project-dependent plant engineering [⁸]. Each module is designed once, including its physical design of the process step and the creation of the information technology interface to higher level systems [²²]. The function of a module remains generic for various applications [²²], whereas plant engineering focuses on integrating the MTP into the POL, orchestrating the services, and implementing project-specific requirements [^{8, 9}]. This significantly reduces engineering efforts for MP construction [^{9, 22}].

Researchers have presented various approaches to planning MPs in recent years, but almost exclusively on process engineering production. In [³³], the authors present a modul-based planning approach for planning new MPs in a structured and time-efficient manner. Researchers in [^{34, 35}] present structured approaches for the selection and configuration of PEAs for process design. The authors in [³⁶] use a multi-criteria algorithm to select equipment modules for flexible, modular production plants with low investment costs. In [¹⁸], the authors present an approach for selecting modules from a database. The researchers in [³⁷] consider the automated creation of an MTP skeleton from existing planning data, which they create during the engineering of a PEA.

In modular process plants, engineers reuse information and pre-engineered modules more frequently [⁸], accelerating engineering phases and reducing effort for final plant design [^{9, 18, 35}]. Modularization has proven to lower planning effort, shorten time-to-market, and an increase flexibility [⁸]. The F³Factory project demonstrates a 25 % reduction in design effort [³⁸], whereas another pilot application reduced engineering and commissioning efforts by more than 50 % [^{8, 39}]. A case study indicates that a holistic rollout of modular concepts can further cut time-to-market for pharmaceutical products by an additional 50 % [⁴⁰].

In the field of logistics, researchers have not identified any approaches that explicitly address the planning of modular production logistics plants in the process industry, though conventional logistics planning is time-intensive [⁵]. Although modularization has shown efficiency gains in production, its potential in logistics remains untapped. The need for digital planning support tools has been highlighted in the previous research [⁹], yet existing methods do not address the specific challenges of (re-)configuring and evaluating modular logistics plants.

3.1 Requirements for an Accelerated Planning Process

The conventional planning of logistics in and around production plants in the process industry is characterized by time-consuming and partly manually performed planning efforts [⁵]. In addition, the supply and disposal of production plants that do not follow a modular concept are initially designed for a defined operating point and do not optimally meet current demand. This means that plants already operate in the ramp-up phase under the conditions of an operating point of maximum output, some of which is only in the distant future. Short-term adjustments to changed market conditions are difficult to implement due to lengthy and partly manually executed planning activities. Thus, each planning procedure has unique characteristics to a certain extent. Another problem is that the know-how in conventional planning processes is often personalized. In order to use the potential of modularization for production logistics plants, the procedure for planning MPs must meet various requirements.

First, researchers and developers must significantly accelerate the planning process to respond to dynamically changing situations such as frequently changing types of products, packaging units, or quantities. In order to enable structural changes such as numbering up or down or a step-by-step ramp-up, the planning of MPs must become faster. This aligns with the need to automate planning: on the one hand, because the frequency of (re-)configurations will increase and thus the absolute planning effort will rise, and on the other hand, to minimize the risk of error-prone planning. The reuse of planning content is another goal of the new planning process in order to reduce repetitive tasks. In addition, the planning process must be digitally feasible to support the planner. This means that planners can safely carry out (re-)configuration planning for MPs even with little planning experience.

For the accelerated planning of modular logistics plants, the logistics modules themselves must meet key requirements for transformability. These enablers—universality, modularity, compatibility, mobility, and scalability—are essential to ensure rapid (re-)configuration and integration within dynamic production environments [⁴¹]. Universality ensures that logistics modules can handle varying product types and packaging requirements without extensive modifications. Modularity allows for the flexible addition, removal, or exchange of functional units, supporting stepwise capacity adjustments. Compatibility enables seamless integration of modules into existing production and logistics infrastructures through standardized interfaces. Mobility simplifies the physical relocation of logistics units to align with changing operational demands, whereas scalability ensures that the system can be adapted to varying throughput requirements without major planning disruptions. Section 4.2 details how these enablers are incorporated into the conceptual design of logistics modules to ensure efficient and flexible planning.

3.2 Concept of an Accelerated Planning Process

One of the main advantages of modular production logistics is that planners can scale the system performance by numbering up or down very easily [⁷] and replace individual modules—for example, for maintenance and cleaning activities—very quickly. For the operator of a corresponding plant, this means that a plant does not have to be initially designed for a maximum output but can be reconfigured in case of comparatively short-term changes in the expected system output. If such a need for change arises, the operator would plan the exact production quantities required, determine requirements for reconfiguration, and then order any missing logistics modules from a manufacturer or service provider. After delivery and—compared to today—significantly accelerated implementation, the reconfiguration would be complete.

Fig. 1 visualizes the methodology of the accelerated planning process in the form of a decision tree. The planning process consists of the phases of modeling as well as simulation and evaluation, which partly run iteratively.

Preliminary plausibility checks help to identify discrepancies such as missing transport routes, undersized storage facilities, or excessive costs. These checks are carried out iteratively each time a module is added or changed. Planning benefits from the possibility of carrying out preliminary tests quickly, as they do not require a simulation. If the model does not meet the required target values, the planner returns to an earlier phase of the modeling process to adjust module parameters, select alternative modules, or modify target values before finalizing the model and transferring it to the model repository.

The second step of the planning process includes the simulation of the modeled configuration and its evaluation and comparison with alternatives. As a result of the simulation, the planner receives information about the compliance or non-compliance with the target values on the one hand and logistical and economical key figures on the other hand to compare these with the key figures of other alternatives. The simulation generates key indicators of individual modules like capacity utilization, waiting time, idle time, downtime, and maintenance time as well as key indicators of the overall system such as throughput, stock levels, lead times, or adherence to schedules. The planner can retrieve the result reports of each simulation run for comparisons even after reconfiguration of the plant. Based on these evaluation reports, the planner decides whether a reconfiguration is necessary. For example, if orders are completed significantly after the planned completion date, the planner can reconfigure the system by changing the module parameters, malfunction data, and the order load. The planning process concludes when the system processes the order load satisfactorily.

On the basis of these decision criteria, the planner can derive further adaptations for disturbance correction. Furthermore, the planner can optimize the maintenance intervals by creating individual maintenance plans for the modules. The simulation with one or more switch-off modules and evaluation according to planned delivery dates or maintenance intervals allows the planner to decide whether a replacement module is necessary or not.

Compared to conventional logistics planning in the process industry, the workflow shown in Fig. 1 introduces significant efficiency gains through modularization, with additional steps highlighted in red. In the modeling phase, modular planning enables the direct selection and parameterization of predefined modules from a standardized module library, eliminating the need for individual design. This reduces engineering time, as planners work with pre-validated components rather than designing each unit individually. Additionally, modular planning leverages a structured model repository, allowing planners to reuse and adapt existing models instead of relying on personal expertise, which in conventional planning is often unstructured and not easily transferable. Conventional approaches also require extensive manual data management, slowing down decision-making and increasing planning effort.

The introduction of reconfiguration as a distinct planning case further accelerates the process. In modular planning, adjustments can be made by modifying existing configurations, significantly reducing planning time. In contrast, conventional planning treats reconfiguration as a full redesign, because previous planning efforts often cannot be directly reused. This results in longer lead times and higher costs for retrofitting. Additionally, modular systems standardize interfaces and compatibility, enabling “Plug & Operate” adjustments, whereas conventional systems require complex engineering to integrate new components.

Simulation and evaluation also become more efficient with modular planning. If simulation results indicate non-compliance with target criteria, alternative models can be generated quickly or retrieved from the repository without restarting the process. This enables fast iterations and optimization with minimal delay. In contrast, conventional planning often lacks integrated simulation at early stages, leading to late-stage revisions that require significant rework. In the case of operational disturbances, modular planning allows planners to quickly assess the impact of replacement modules and implement solutions with minimal downtime. Conventional planning, by contrast, lacks systematic fallback options, requiring time-consuming troubleshooting and manual adjustments, which lead to extended disruptions.

4.1 Requirements for a Planning Support System

According to the accelerated planning process and its requirements outlined in Section 3, the planning support system requires that a planner can (re-)configure an MP in an accelerated, digital, and partially automated manner with partially reusable planning content. The definition of these requirements is based on structured expert interviews with practitioners from the process industry, including plant planners and engineering professionals. Their insights reflect the specific challenges encountered in planning and adapting MPs efficiently.

These requirements ensure that the planning support system effectively addresses the challenges identified by industry professionals and aligns with the principles of modularization and digitalization in modern plant engineering.

4.2 Module Library and Conception of Logistics Modules

Although the module library considers the production modules as a black box in order to reduce complexity and the focus of this article, the logistics modules are dealt with in detail. However, the analysis of the state of the art reveals a lack of availability of modular, “Plug & Operate”-capable production logistics plants [^{9, 11}]. The work of Kaczmarek et al. and Hachmann et al. supports this analysis and demonstrates the necessity of the conception of further logistics modules in modular chemical production [^{12, 16}]. Therefore, as a further preparatory step for validating the accelerated planning process and thus the digital planning support system, we will conceptualize and specify missing logistics modules for the functions of filling, palletizing, packaging, and load securing. The reference process model presented in [^{28, 29}] for the design of flexible intralogistics units for production supply and disposal is used as a basis for this, with which logistical tasks can be identified and corresponding module sections can be carried out.

Particular attention is paid to the transformation enablers according to Wiendahl et al. [⁴¹], which ensures that the logistics modules are designed to meet key adaptability enablers such as universality, modularity, compatibility, mobility, and scalability. These enablers are crucial for ensuring that the logistics modules can function effectively in a dynamic, modular production environment.

The logistics modules must be adaptable to various products and packaging types, enabling them to handle goods with differing physical properties or packaging requirements. This adaptability is essential for responding quickly to changing production or market demands. Furthermore, the modules should be designed as independent, interchangeable units that can be easily reconfigured or scaled to accommodate different production needs. Each module must be capable of operating autonomously while integrating seamlessly into the broader modular system. In addition to their modular design, the logistics modules must be compatible with various production systems, both existing and future. This includes ensuring standardized interfaces and communication protocols that enable smooth integration with other modules and control systems. Mobility is also a key consideration, as the modules need to be easily transportable to different production locations. Standardizing their physical size, for example, using ISO containers, allows for quick deployment in new areas, ensuring flexibility in adapting to changes in the market. Finally, scalability is an important factor, as the logistics modules must be able to expand or contract depending on fluctuations in production volume or demand. This scalability ensures optimal resource utilization and prevents inefficiencies, such as underutilized capacity or bottlenecks in the logistics process.

Based on existing plants available on the market today, it will be determined at which points technologically reasonable cuts for modularization can be made, so the different plant components can be installed in suitable container formats (ISO standard container, 10-, 20-, or 40-ft). The individual containers will then be assembled to form the complete plant. At this point, reference is made to the work of Jürgensmeyer et al. They describe a transformation model for MPs and demonstrate that the path from conventional production plants to modular, smart production environments can be made step-by-step [¹⁹].

In order to prepare these modules for the use in the planning support system, they are described with regards to the transformation step by considering incoming and outgoing goods and are specified by defining simulation-relevant parameters such as the typical lead time and the throughput. In addition, the specification of the individual modules includes further parameters such as the secondary packaging materials and the space required.

To reduce complexity, we conclude the following restrictions. On the one hand, the information regarding throughput is based on existing plants available today. We can only estimate possible influences on the performance of such plants as a result of their modularization. In order to take the idea of modularization into account, we divide the individual plants into ISO containers (20-ft). The container dimensions determine the required space. Furthermore, we do not consider country-specific requirements for the individual modules.

Fig. 2a depicts an example of the conceptual design for one of these logistics modules, a modular stretch hood machine. It shows the module partition, which divides the machine into three parts. Although the first module (20-ft container) contains the foil feeding and the switch cabinet, the second module (20-ft container) contains the roller conveyors and the lower part of the foil overlayer (frame). The third module (10-ft container) on top of the first two modules contains the upper part of the foil overlayer. Fig. 2b depicts the modular stretch hood machine in container design.

4.3 Simulation-Based Planning Support System

The planning support system contains various services for the planning and evaluation of MPs. In this way, different planning and evaluation procedures can be encapsulated as services, which can be used individually according to requirements. The planning tasks of the planning support system include, for example, resource and demand planning for forecasted or existing orders, short-term procurement and production planning, forecasting of future inventory developments as well as checking throughput, dimensioning, and performance limits. In order to fulfill these planning and evaluation tasks, the planning support system integrates with a discrete-event simulation environment, enabling the execution of simulations of the virtual production plants. The architecture of the planning support system corresponds to a client-server architecture. Therefore, the planning support system uses an “AngularJS Business-Logic” framework on the front-end side and a “Spring Boot RESTful Web-Services” framework on the back-end side. Additionally, for evaluating modularly structured plants, the planning support system interfaces with the INOSIM software suite for material flow simulation, a simulator for batch and mixed batch continuous processing systems. Via this interface, the user automatically transfers the design parameters of the considered plant to a simulation model to perform detailed quantitative analyses on the complete plant (covering both the production process itself as well as the connected modular logistics system), including, e.g., bottleneck analysis and debottlenecking, what-if-analyses, or production scheduling [⁴²]. Section 4.4 demonstrates the functionality of simulating mixed batch production schedules and bottleneck analyses with an example.

Further services of the planning support system include the previously presented module library consisting of production and logistics modules and a graphic modeling surface for the configuration and parameterization of modular systems. Additionally, an essential feature of the planning support system includes services for persisting and loading plant models and automatically generating simulation models using the integrated discrete-event simulation engine. In order to meet the goal of a simulation-based evaluation of modular production based on key figures and decision support, the planning support system includes a service for the evaluation of simulations based on suitable key figures and their graphical representation. Furthermore, the planner can manage modules and simulation runs using different views.

4.4 Exemplary Production Scheduling and Bottleneck Analysis

The following example considers the filling of pesticides in 1-, 2-, and 5-L canisters with subsequent weighing, canister labeling, and carton packaging. Fig. 3 shows the corresponding process chain.

The digital planning support system provides—as a core functionality—the simulation of production schedules. Especially in the production of several product variants in one system—e.g., the filling of process-engineered products in different types of containers—product changes and the resulting different machine throughputs create a challenge to develop reliable production schedules and meet delivery dates. By simulating the production schedule, the planner can identify possible capacity bottlenecks or overcapacities, revealing the necessity of reconfiguration measures.

Therefore, Fig. 4 shows a detailed production schedule including the degree of utilization of the logistics modules for an exemplary production scenario of 1-, 2-, and 5-L canisters. The exemplary schedule illustrates that for this scenario, both the weighing and labeling modules (lines 5 and 6 in the diagram) create bottlenecks in production for this canister size mix. In contrast, the modules for depalletizing, filling, and packaging experience idle times.

5.1 Hypotheses on the Effects of Modularization on Production Logistics

5.2 Testing the Planning Support System

The evaluation uses a test scenario that includes selected value-added sections of the process industry, for which the use of modular logistics plants is likely. For example, modular logistics plants are a promising approach for the filling of chemical products in the construction industry. In the test scenario, we consider the filling of cement into 25 kg bags with subsequent palletizing and load securing. We examine the influence of a fluctuating order volume due to a volatile market demand on the necessity of a reconfiguration of the system. Fig. 5 shows the corresponding process chain as well as the plant layout. In the initial situation, the system contains one filling module. We assume that the cement to be filled represents a resource, which is always available, so the silo (“incoming goods”) never runs empty. The aim is to estimate the critical order load of the system and to identify capacity bottlenecks. Therefore, we examine the planning objectives “order backlog” and “plant utilization” in this test scenario. The simulation in this test scenario uses data based on a detailed and real order load, which is an input variable of the planning support system. The order load extends over a period of 6 months and thus represents a sufficiently representative period of time.

The evaluation of the simulation results generated via the integrated discrete-event simulation engine in the initial situation shows that the filling module is working at 100 % capacity, and the order backlog, shown in Fig. 7, builds up continuously and cannot be processed. Therefore, the filling module represents a capacity bottleneck in the initial situation, which necessitates the reconfiguration of the system. Following the workflow mapped in the UML diagram (see Section 3.2), the planner first analyzes the simulation results in comparison to the planned target values, such as delivery deadlines, throughput rates, production costs, and maximum allowable inventory levels. Observing that the results do not meet the planned targets, the planner identifies the filling module as a bottleneck. This leads to a return to the modeling phase, where the planner modifies the configuration by integrating a second filling module with the same capacity profile. The reconfigured model is saved as a model alternative before the planner proceeds to the simulation step again, loading the new configuration and running the simulation under the same conditions. Fig. 6 shows the process chain and the plant layout of the reconfigured system.

The renewed evaluation of the simulation results shows that the utilization of the downstream palletizing and stretch hood modules approximately doubles compared to the initial situation, see Fig. 8. This occurs because we supply these modules with material from two filling modules after we reconfigure the system. This aligns—according to hypothesis 1—with the expectation that numbering-up in consequence of modularization leads to an increase in capacity utilization. The capacity utilization of the two filling modules is 100 % each, i.e., the total capacity remains insufficient to complete all orders on schedule. However, the reconfiguration positively affects the order backlog, as shown in Fig. 7. The average order backlog reduces significantly, and the maximum order backlog also decreases by more than 50 % by using a second filling module. Following the iterative workflow, the planner once again compares the results to the target values and determines that the filling process remains a bottleneck. Consequently, the planner would repeat the process, adding a third filling module, re-running the simulation, and evaluating the impact until an optimal configuration is achieved.

5.3 Summary of Simulation Results and Comparison to Hypotheses

The presented test scenario investigates the effects of a fluctuating order load on a modularized system. The analysis of the plant utilization shows that the filling module operates at 100 % in the initial situation and thus represents a capacity bottleneck, whereas the palletizing module and the stretch hood module are significantly less utilized. Reconfiguring the system by adding a module to the bottleneck process shows that the capacity of the bottleneck process specifically increases, and the utilization of all other modules also increases. In a non-modular setting, we would duplicate the entire system, thus unnecessarily multiplying the capacity of processes not operating at full capacity. The test scenario demonstrates that doubling the bottleneck capacity improves the utilization of other processes. Thus, the simulation results from the test scenario confirm the first hypothesis, “modularization increases plant utilization,” in the analyzed case.

In addition, the test scenario investigates the effects of numbering-up in the course of modularization on the order backlog. The simulation results show that reconfiguring the system by adding a module to the bottleneck process has a positive impact on the order backlog. The maximum order backlog decreases by more than half, and the average order backlog also decreases significantly. By making plant capacity more flexible through modularization, we can react to fluctuating market demands. As a result, the system processes the order backlog more quickly, as shown in the test scenario. Overall, this means that we confirm the second hypothesis “modularization reduces the backlog of orders” in the case of the test scenario.

In conclusion, the evaluation shows that the planning support system fully maps the required main functions described in Section 4.1. The planning support system thus offers the possibility of planning, (re-)configuration, and evaluation of modular logistics plants with high usability and graphical support. The validation of the correct functioning of the planning system and the simulation-based results show that the concept of the accelerated planning process meets the requirements for the planning procedure for MPs in the process industry.

However, as introduced in Section 4.4, the functionality for production planning and bottleneck analysis still requires further development to address fluctuating demand and sequence planning in modular process industry setups. Addressing these gaps, recent research highlights relevant approaches. The authors of [⁴³] investigate batch production scheduling using mixed-integer linear programming, proposing a multi-step decomposition approach to enhance computational efficiency—a concept that could be extended to modular production environments. The work of [⁴⁴] integrates sustainability aspects and production variability in multi-level chemical processes, offering insights into balancing economic, environmental, and operational trade-offs under uncertainty. Lastly, [⁴⁵] focuses on lot-sizing challenges in the chemical industry, considering parallel processes, inventory constraints, and production flexibility, all of which are critical factors in managing fluctuating inputs. These studies provide a foundation for expanding the planning support system to incorporate demand-driven sequence planning and adaptive scheduling mechanisms.