APM and AIP Integration for Joint Optimization of Productivity and Reliability Using Simulation Experiments

Abstract

This study presents a methodology for integrating Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms for joint optimization of productivity and reliability using simulation experiments. This research combines data from an APM, which provides information on equipment reliability, and a simulation module of AIP software that offers detailed technical data of a set of alternative equipment in a sort of catalog. Criteria such as availability, reliability, criticality, and utilization levels, based on historic stored data, are used to evaluate different equipment configurations. Such data are provided by the APM platform, while the productivity and efficiency of existing and candidate equipment are captured from the configuration module on the AIP platform. Key aspects of this work point to the possibility of applying it in two main stages of a system’s life cycle: the design stage, where the project is in its conceptual design phase, and the operation or exploitation stage, where newer configurations are considered to prioritize operational adjustments and optimizations due to the inherent constraints of reliability and maintainability aspects. Through the development of a specifically developed model, which is applied to ensure optimal selection of equipment and configurations, it is possible to obtain new equipment configurations that ensure operational continuity and efficient production performance without exceeding budgetary restrictions and energy consumption limits or compromising productivity.

1. Introduction

In the mining industry, equipment reliability is one of the fundamental pillars to ensuring the continuous and efficient operation of processes, especially in key processes such as comminution or material size reduction, where operational performance largely depends on the proper selection of equipment. Decisions related to equipment configuration can critically affect productivity, downtime, and associated costs, making optimization in configuration selection essential for the success of mining operations. Without proper selection, the frequency of failures or underutilization of equipment can increase, resulting in unplanned downtime and rising operational costs. Addressing this issue from a reliability-based approach is crucial to ensure operational continuity and sustainability [1].

There are several definitions of reliability quoted in the literature, but the one most often stated corresponds to the probability that a system performs its intended function without failure for a specified period of time under stated operating conditions [2]. In a mining environment, these principles are crucially applied to equipment and processes. Equipment configurations must meet not only production demands but also reliability standards that ensure continuity over time, minimizing the impact of failures on productivity [3]. However, achieving this balance between productivity and reliability is not straightforward, as the most productive configurations are not always the most reliable, and vice versa. To enhance reliability, systems frequently include redundant components—additional elements designed to take over in case of a failure. Although this approach improves system resilience, it can also lead to reduced overall efficiency and productivity by introducing greater complexity and incurring higher upfront costs. This highlights the need for methodologies that efficiently integrate both aspects.

Asset Investment Planning (AIP) solutions enable organizations to rethink asset management strategies beyond simple replacement decisions, especially in environments characterized by constrained resources and increasingly complex revenue structures. By integrating data from Enterprise Asset Management (EAM) and Asset Performance Management (APM) systems—particularly asset health, criticality, and depreciation—AIP tools support the evaluation of multiple investment alternatives throughout the asset life cycle. These systems facilitate objective, data-driven decision-making aimed at optimizing costs and minimizing operational risks, encompassing everything from failure prediction to the planning of maintenance interventions, asset reallocation, or replacement. AIP platforms are intended to align investment decisions with organizational strategies, incorporating economic analysis under constraints such as budget limitations and regulatory standards (e.g., ISO 55000 [4]). To address the complexity of comparing heterogeneous investment proposals, AIP platforms generate customized value functions that translate strategic goals into quantifiable, weighted metrics. However, the literature still lacks a structured approach for effectively integrating these platforms with those that provide the data required to feed risk assessment models (APMs)—particularly information related to the reliability and maintainability of physical assets to consistently evaluate and prioritize investment options [5].

The main research question of this study is whether the integration of Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms—supported by a middleware algorithm—can effectively guide the selection of equipment configurations that optimize productivity, reliability, and energy efficiency in industrial systems. The working hypothesis is that creating a feedback loop between real-time operational data from APM and the long-term planning capabilities of AIP allows for identifying configurations that perform better than those selected by traditional methods. This hypothesis is tested using simulation techniques and reliability analysis, particularly through RAM (Reliability, Availability, Maintainability) models.

The integration of Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms is a technically demanding yet strategically essential process to align operational asset condition with long-term financial planning. APM systems enable real-time monitoring, failure prediction, and reliability analytics, while AIP platforms focus on capital planning, investment prioritization, and life cycle modeling. Bridging these platforms facilitates data-driven investment decisions based on real-time asset health and risk assessments.

Key integration challenges include data heterogeneity, semantic misalignment, and differences in data granularity and latency. APM generates high-frequency data from sensors and maintenance logs, whereas AIP relies on aggregated, structured information over extended horizons. Middleware solutions address these differences through ETL pipelines, API gateways, message brokers, and data virtualization tools. Technologies such as Talend, Apache NiFi, RESTful APIs, Kafka, and time-series database connectors (e.g., OSIsoft PI, InfluxDB) play a critical role in enabling secure and scalable integration.

Typical data exchange patterns include scheduled batch transfers, event-driven messaging, and API-based orchestration for real-time scenario evaluation. Core data structures transferred from APM to AIP include asset health indices, remaining useful life (RUL), failure probability curves, maintenance history, and condition monitoring trends—each supporting risk-based investment modeling.

Ensuring semantic alignment across platforms requires consistent asset identifiers and taxonomies, typically enforced using ISO 14224 [6] or IEC 81346 [7] standards, and supported by master data management (MDM) tools. Data governance is also critical, relying on role-based access control (RBAC), secure authentication (OAuth2, OpenID Connect), and traceable audit trails, especially in regulated or critical infrastructure contexts.

This paper presents the development of a decision-support framework for selecting optimal equipment configurations aimed at maximizing operational productivity through enhanced system reliability. The proposed framework leverages the integration of two technological platforms: an Asset Performance Management system, which supplies critical data on equipment reliability, and an Asset Investment Planning tool, which offers detailed technical and operational specifications for a range of equipment alternatives sourced from a predefined catalog. To enable seamless interaction between these platforms, a middleware algorithm was developed. This algorithm systematically evaluates multiple configuration scenarios and identifies optimal equipment combinations that satisfy predefined reliability thresholds and productivity requirements.

Furthermore, a key aspect of the proposed framework is the consideration of the energy consumption of the selected equipment. This allows the generation of configurations that not only ensure operational continuity but also optimize energy use without compromising productivity. This is particularly relevant in the mining industry, where energy efficiency is a key factor for cost reduction and operational sustainability. Ultimately, this work contributes to improving decision-making in equipment management and developing a more strategic approach to their configuration, considering not only productivity but also reliability and energy efficiency.

2. Theorical Background

In the current reality, marked by the challenges of a VUCA (Volatile, Uncertain, Complex, and Ambiguous [8]) world and global disruptions affecting production systems (such as health, geopolitical, and climate-related threats), the need arises for implementing reliable and robust systems which can adapt quickly to operational changes and shifting asset conditions, ensuring and improving resilience and sustained performance [8]. Improving an industrial plant’s performance relies on several key factors, including the effective execution of preventive maintenance (PM), continuous asset condition monitoring, and comprehensive Asset Performance Management (APM) and Optimization (APO) [9].

In the management of industrial assets, two critical concepts emerge that aim to optimize both operational and strategic and long-term oriented performance: Asset Performance Optimization (APO) and Asset Investment Planning (AIP). Although these two concepts are closely related, they have distinct purposes, methods, and outcomes that cater to different aspects of asset management. One of the key purposes of APM is to reduce the risk of unexpected failures, especially in critical systems. Therefore, it is paramount to focus precisely on this by aiming to minimize failure risk through dynamic decisions that consider both asset performance and the operational context [10]. Also, APM focuses on improving the operational performance of assets mainly linked to reliable equipment and processes [11]. AIP, essential for improving productivity and reducing risks in industrial systems, should align its results with capital expenditure (CAPEX) and operational expenditure (OPEX) optimization throughout the asset life cycle. Both approaches, jointly, support more sustainable financial and operational asset management [12]. To enhance strategic decision-making in asset management (AM), it is crucial to integrate APM with AIP.

Asset Investment Planning platforms constitute a strategic tool that helps companies make complex decisions about budget allocation and long-term asset management. Unlike APM and Enterprise Asset Management (EAM) systems, which focus on tactics and operations mainly related to assets’ availability, AIP addresses tactical and strategic decisions mainly focused on capital expenditure (CAPEX) [13] and the composition of the proposed alternative configurations. Although initially adopted in regulated sectors such as energy and water utilities, AIP is rapidly expanding into industries like oil and gas, transportation, and manufacturing due to its ability to optimize investment decisions and enhance long-term financial performance [13,14].

Asset Performance Management is primarily concerned with the optimization of assets’ operational performance [15]. The primary goal of APM is to maximize the availability, reliability, utilization, and efficiency of assets through tools like reliability engineering, simulation and sensibilization (RAM analysis). APM can be seen as a tactical approach to asset management because it addresses the short- and medium-term performance and operational challenges faced by assets in an organization [16].

In APM, the focus is on continuously improving how assets perform within their respective operational environment. The goal is to reduce unplanned downtime, extend the useful life of assets, and ensure that assets are performing at their best, which, in turn, reduces maintenance costs. Naidu et al. propose a framework for maintenance decision-making in the power sector that integrates traditional practices with data-driven methods by combining business pull and technology push strategies to optimize RAM performance and align innovation with operational needs [17].

In a certain manner, the APM approach is dynamic, with operational decisions, even on Multi-Assets scenarios, ensuring that the right maintenance actions are carried out at the right time [18]. Finally, in some cases, legal, normative, or regulatory factors have to be regarded in such APM-based analysis [19]. Moreover, APM is increasingly tied to data analytics, with big data being leveraged to predict failures [20], optimize processes, and enhance operational productivity and throughput [21] through Machine Learning techniques. According to Firstantara et al. [22], recent studies on Asset Performance Management (APM) have introduced advanced data-driven techniques, such as machine learning, which already demonstrate significant potential for managing high-complexity assets, enhancing availability, reducing costs, and supporting sustainability. However, these studies are primarily focused on contexts with robust resources and sophisticated infrastructure, leaving a gap in understanding how such practices can be adapted and optimized in organizations with limited resources. Also, some visualization tools are being used as a means to enhance APM’s functionalities [23].

On the other hand, Asset Investment Planning deals with strategic decision-making, mainly related to asset investments, focusing on capital expenditure (CAPEX). AIP concerns itself with the strategic point of view of asset management, primarily aiming to plan and allocate and configure resources efficiently to acquire, add, or replace assets in alignment with long-term business goals. In other words, it supports long-term planning by evaluating how investments in assets can provide the maximum return on investment (ROI) over time. AIP involves detailed plant modeling and stage analysis to predict the future costs, performances, and outcomes of investments in a risk analysis scenario [24]. Approaches like Life Cycle Costing (LCC) are commonly used in AIP to complement the assessment of the total cost of owning and operating an asset throughout its life, including acquisition and reconfiguration costs [25]. The role of AIP is to make or assist investment decisions that align with the organization’s strategic objectives, balancing financial opportunities and constraints with the operational needs of the business [26]. The AIP approach may be used to determine when and how to invest in new assets, whether through replacement, upgrades, or new acquisitions [27]. Recently, Industry 4.0 tools have been included as strong techniques to develop adaptative solutions in assets management; therefore, Intelligent Asset Management Models (IAMMs) support long-term planning by simplifying data complexity and aligning with advanced maintenance strategies [28]. A recent approach gaining momentum in physical asset management is the use of Digital Twins [29]. This technology constitutes a powerful tool for analyzing and optimizing production systems by not only enhancing efficiency but also focusing on asset availability and throughputability. Digital Twins enable real-time monitoring, simulation, and performance forecasting, making them essential for improving operational resilience and maximizing system productivity [13,14].

Despite their differences, APM and AIP are complementary. The integration between an APM and an AIP is deemed both essential and valuable, hence the need for these platforms to work in an integrated manner [5]. AIP ensures that the right assets are being acquired and maintained at the right times and costs, while APM ensures that assets are operating efficiently, which directly impacts their useful life and performance. Both strategies must work together to ensure the overall success of asset management. As Figure 1 depicts, the insights gained from APM can inform AIP decisions. If APM reveals that certain assets are consistently underperforming or reaching the end of their useful life, AIP can then decide on the need for investment in replacements or upgrades. Similarly, AIP decisions, such as whether to invest in new technologies, can lead to changes in the APM approach to accommodate the capabilities of newer or more advanced assets, mainly from the reliability point of view.

Analyzing different stages in asset investment planning provides a transparent and reproducible basis for evaluating desirable productivity levels and associated costs. This approach supports long-term cost management through strategic infrastructure investments while enabling identification and mitigation [30]. Given the critical role of asset risk management in production systems, future research should focus on enhancing these evaluation methods [31].

Selected studies indicate that economic performance is closely linked to a tactical approach in implementing APM. This adoption is often driven by the potential to reduce maintenance-related costs [32,33] or, more broadly, by the need to increase economic returns [34].

However, customers often have extensive needs and high expectations, which can surpass the capabilities or available resources of individual solution providers, complicating the delivery process. Therefore, integrating APM and AIP platforms is essential for maximizing asset management efficiency. Their integration enables AIP decisions to be supported by APM’s operational insights, allowing for stage testing and sensitivity analysis. This approach provides more realistic investment evaluations by considering both financial and operational impacts, such as asset availability and throughput capacity. As a result, organizations can align investment strategies with operational goals more effectively.

The selection and optimization of equipment in mining operations is a critical process that directly influences efficiency and operational costs. Various methodologies have been developed to address this challenge, with the Life Cycle Cost (LCC) analysis being particularly notable [35]. Such analysis evaluates the total cost of ownership of equipment, including acquisition, operation, maintenance, and final disposal. A previous study highlights the importance of using simulation models to assess LCC and make informed decisions in machinery selection.

Digitalization has significantly transformed this process, providing advanced tools such as simulation software and performance analysis platforms that allow for comprehensive system evaluation both before and after project implementation. By modeling different combinations of equipment, it is possible to assess their performance under specific conditions, optimizing parameters such as reliability, availability, and energy consumption. This significantly reduces the risks associated with equipment selection, as decisions are based on realistic simulations and precise technical data.

This paper proposes an integration framework of AIP and Asset Performance Optimization (APO) platforms to enhance decision-making in mineral crushing plants. This integration, and its corresponding middleware, would enable comprehensive analyses by linking long-term investment strategies with actual asset performance data. By combining AIP’s strategic planning capabilities with APO’s operational insights, it is possible to simulate various operational stages, assess equipment reliability, and evaluate potential upgrades or replacements. Overall, the proposed approach, enhanced by digital tools, focuses on ensuring the durability and stability of plant size and capacity analysis projects. Additionally, probabilistic RAM analyses and Monte Carlo simulations are used to evaluate and optimize considering system reliability [36,37]. This allows for quantifying the increase in system availability through greater overall reliability and lesser risk of critical failures, ensuring long-term operational continuity, runtime, and higher throughput.

This approach would improve the accuracy of investment decisions by considering factors such as equipment availability, throughput efficiency, and life cycle costs, ultimately maximizing the plant’s productivity and investment returns.

3. Main Methodology

As previously mentioned, the scope of this research is defined by the need to integrate, through a middleware algorithm, two asset management platforms to support the selection and optimization of equipment configurations in industrial plants. The proper selection and arrangement of equipment is a critical factor in enhancing operational efficiency within production systems. This optimization process relies on multiple criteria to identify the equipment that has the most significant influence on the overall system reliability, enabling more targeted and effective decision-making. Those criteria are briefly described below:

• Availability and Utilization: These criteria are incorporated using data provided by the APM platform, which is obtained either from actual operational databases or generated through Monte Carlo simulations conducted within the same environment. These simulations use performance and effectiveness metrics to evaluate the behavior of equipment under varying conditions. Availability is defined as the percentage of time the system is capable of operating or producing, while utilization—also referred to as the service factor—represents the proportion of time an asset is actively operating within a given period. Both parameters are fundamental to ensuring the continuity and efficiency of operations, and their application in this study ensures the methodological alignment with the proposal.
• Criticality: The identification of critical assets is a fundamental concept in reliability engineering, referring to the level of risk or operational impact associated with the failure of a specific piece of equipment. It serves as a basis for identifying and prioritizing assets whose malfunction could significantly compromise system performance or operational continuity. This criterion is essential for directing improvement efforts toward the most vulnerable components, enabling targeted decisions regarding optimization, maintenance, or replacement. Equipment criticality can be assessed through methods such as Failure Modes and Effects Analysis (FMEA) or Fault Tree Analysis (FTA) [38,39], both of which help pinpoint weak points in the system and allocate maintenance resources more effectively. These analytical tools are integral to the capabilities offered by APM platforms.
• Reliability: A key concept in maintenance engineering and industrial asset management. It is defined as the probability that an item can perform its required function during an established time interval and under defined conditions of use. Reliability theory is based on probabilistic models that evaluate the behavior of equipment over time, considering factors such as wear, failure criticality, and failure periodicity, usually expressed by the Mean Time Between Failures (MTBF). These models allow for estimating the capacity of equipment to meet operational goals and ensure operational efficiency [40]. It is an essential criterion in configuration/equipment selection. This criterion integrates the other aspects and provides a solid framework for evaluating the robustness of the configurations. Reliability analysis and optimization models are applied to evaluate different configurations, where equipment can be arranged in series, parallel, or split, which directly impacts operational redundancy and load distribution among equipment. For example, parallel systems theory allows for the distribution of operational load across multiple pieces of equipment, thereby reducing the criticality of any single piece and improving overall reliability [41].
• Energy Efficiency: As mentioned earlier, energy efficiency is a crucial factor in selecting configurations in heavy-duty operations. The energy consumed by, for instance, crushing equipment can represent a significant cost in the operation, making it necessary to consider energy consumption in optimization models. Energy management theory suggests that equipment configurations should seek a balance between productivity and efficiency, minimizing energy consumption without compromising operational reliability [42]. The middleware developed to connect APM and AIP platforms integrates this constraint directly into the optimization loop: any new equipment configuration must not exceed the baseline system’s total energy consumption. This constraint is evaluated alongside other operational requirements such as minimum reliability thresholds (e.g., MTBF), production capacity targets, and availability or utilization metrics. Total energy demand is computed as the sum of each unit’s power draw multiplied by its simulated operating time, and only those configurations that satisfy this condition proceed to reliability simulation and evaluation.

The project optimization can be seen from two main perspectives or in two different scenarios: the project during the conceptual development stage and the operation of an already functional plant (Figure 2). This distinction allows the proposed methodology to be adapted to the specific conditions and constraints of each stage, maximizing its applicability and effectiveness.

3.1. Projects in Conceptual Development

At this stage of the asset life cycle, the proposed framework facilitates the evaluation of multiple equipment configuration alternatives with minimal cost and effort, enabling a comprehensive Life Cycle Cost (LCC) analysis [43]. It allows for the assessment of various configurations based on key criteria such as reliability, productivity, and energy efficiency; the testing of optimal combinations through iterative simulation to ensure alignment with operational, sustainability, and reliability goals; and the proposal of significant design modifications—including the inclusion, replacement, or removal of equipment—without incurring the economic and logistical challenges typically associated with making changes in operational environments. This early-stage application represents the core focus of the research, as it provides the flexibility needed to implement advanced optimization strategies before physical deployment, thus avoiding constraints related to infrastructure, capital expenditure, or operational risk.

3.2. Reconfiguration of Operating Plants

At this stage, the project is already in its operational phase, and the configurations and equipment under analysis are part of an existing, functioning system. The proposed methodology is oriented toward evaluating practical and realistic modification strategies that are adapted to the actual constraints of an operating plant. These strategies may include maintaining well-regulated basic conditions, adhering to appropriate operating procedures, restoring deteriorated components, addressing design weaknesses, and enhancing operational and maintenance competencies. Although the range of modifications is inherently more limited due to the fixed nature of the system, the integration of Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms remains both necessary and valuable, offering structured support for decision-making even within constrained operational contexts. By integrating these two platforms, the following achievements can be attained:

• Evaluate configurations from both an operational and reliability perspective, ensuring that the proposals not only meet production demands but are also sustainable over time.
• Reduce risks associated with unplanned failures by jointly analyzing reliability metrics and technical performance.
• Propose more robust configurations aligned with the project’s operational and energy objectives.

Through a unified view of asset performance and investment priorities, we can achieve greater consistency in evaluating configuration alternatives, the ability to simulate and compare scenarios based on reliability and cost, and more informed, data-driven decisions that align with both operational constraints and long-term strategic objectives.

4. Detailed Procedure

The integration of Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms, along with the execution of the corresponding simulation experiments, can be effectively carried out using a combination of specialized commercial software tools and custom-developed components. For the APM modeling side, reliability, availability, and maintainability (RAM) analyses were conducted using RMES Analytics, a dedicated reliability modeling platform. RMES supports the construction and analysis of Reliability Block Diagrams (RBDs), the definition of failure and repair time distributions such as exponential or Weibull, and the execution of Monte Carlo simulations to evaluate system availability and performance. It also allows for detailed criticality analysis and provides key maintainability metrics, including MTBF (Mean Time Between Failures) and MTTR (Mean Time to Repair).

On the AIP side, equipment configuration and investment planning were modeled through a specialized AIP design module, integrated into a proprietary environment (Bruno by Metso). This module enables the assembly of asset configurations using a catalog-based structure, supporting setups in series, parallel, or fractional arrangements. It also facilitates life cycle cost estimation, energy profiling, and productivity analysis for each configuration.

To link both platforms, a custom Python-based middleware was developed. This middleware automates the extraction of reliability data from the APM system and configuration data from the AIP tool, applies predefined constraints related to performance, energy use, and productivity, and controls the iterative simulation process. Simulations, particularly the Monte Carlo analyses, were executed within RMES and supported by Python 3.13.3 libraries such as NumPy, SciPy, and pandas for data processing, while Matplotlib (v.3.8) and Seaborn (v.0.12.2) were used for visualization. Data were managed through PostgreSQL databases to ensure traceability and integrity, and exploratory analysis and reporting were facilitated through Power BI dashboards and Jupyter Notebooks, enabling clear interpretation of results for decision support. This combination of tools forms a robust computational environment for the integration and joint optimization of APM and AIP platforms in complex industrial contexts.

The process begins with the development of an initial configuration for the project, specifically outlining the equipment involved. This preliminary configuration is established within the design module of the AIP platform and includes a range of available equipment, which can be arranged in series, parallel, or more complex configurations, such as partial redundancies and fractionality-based setups. Within this definition, it is essential to explicitly specify not only the equipment but also the incoming material conditions and the desired output characteristics. This initial configuration results in specific levels of production, energy consumption, and productivity. However, during this first iteration, adjustments or refinements may be required to align the system’s productive performance and energy consumption with the reliability of each constituent component. At the initial configuration phase, key values related to the performance and reliability of each piece of equipment are assigned. These values include metrics such as reliability, availability, criticality, utilization, and failure rates. The data are extracted from the APM platform, considering the respective configurations of the components which are represented by the respective Reliability Block Diagram (RBD) [44]. This platform includes the relevant data on reliability (MTBF); maintainability, frequently expressed by the Mean Time to Repair (MTTR); and failure rates for each piece of equipment, including those selected in the configuration module of the AIP.

Every piece of equipment in the initial configuration is evaluated to conduct a criticality analysis. This is achieved by considering its systemic impact through the propagation of reliability, maintainability, and failure rate values throughout the entire designed system. For this purpose, the RBD is considered. This process will allow the selection of the asset with the highest criticality value. This asset is identified as the most prone to failure or having a negative impact on systemic reliability and productivity, making it the focus of efforts to improve its performance or replace it. The effect on systemic reliability is assessed through a set of RAM-type experiments, in which the probabilistic behavior of the reliability and maintainability performance parameters of each piece of equipment in the configuration under analysis is simulated. The probabilistic behavior of these parameters is stored in the APM analysis module repository and, ideally, should be obtained from historical records of actual equipment in comparable situations or contexts. Another alternative for obtaining these parameters and their probability distribution functions is through literature sources or, lastly, expert judgment. Once the RAM analysis is completed, the results of this analysis are used to identify criticalities.

Once critical equipment has been identified, the previously defined structure within the AIP design module is analyzed to determine potential new configurations. The new configurations are simulated to validate and evaluate their performance, as well as to obtain estimates of autonomy, production, and energy efficiency values. These configurations may include new series, parallel, fractionation, and stockpiling arrangements. The selected equipment, focused on addressing reliability weaknesses, is replaced with new equipment or a more complex configuration. The objective is to ensure that the proposed modifications maintain or improve operational productivity while optimizing system reliability. Additionally, energy consumption is used as a constraint in the searches and recommendations of configurations, so that the new configurations should not exceed the current system’s energy consumption. This not only keeps energy consumption under control but also enhances the plant’s energy efficiency. Eventually, the option of incorporating one or more stockpiles can be considered to strengthen productive capabilities in response to the reliability “weaknesses” detected in the APM module.

This iterative process should be carried out until a termination condition is reached. Such termination condition may occur when the defined configuration meets the requirements of autonomy, productivity, and energy efficiency, and exceeds the minimum level of systemic reliability ensuring compliance with the operational parameters. In addition, the APM system will allow for a more comprehensive and long-term analysis by incorporating variations in parameters and their respective probability distribution functions over an extended period. This involves examining the system from a life cycle cost perspective. The main methodology described is depicted in Figure 3. The colors used for each phase represent the origin of the module utilized in it.

5. Case Studies

To provide more readiness of the proposed methodology and validate its applicability, a case study is presented, focusing on a mineral comminution plant. The study begins by analyzing an existing operational scenario, identifying key reliability issues that impact the plant’s productivity. Subsequently, alternative configuration strategies are evaluated through the application of the proposed methodological framework.

This case study will analyze a size reduction plant that works with granite, initially 800 mm in size, with the aim of optimizing productivity and process reliability. The plant in question is not a real installation but a model created specifically for this simulation. This plant is equipped with various pieces of equipment, including crushers, feeders, screens, a stockpile, stockfeed, and a silo. Through modifications in configuration and the pieces of equipment that constitute the plant, we intend to demonstrate how these changes can influence the efficiency and operational robustness of the plant.

Originally, the plant processes granite to obtain four final products with sizes of 5 mm, 38 mm, 18 mm, and 25 mm. The analysis will focus on how changes in configurations and equipment can improve both productivity and process reliability. Three different configurations will be tested to evaluate the optimization of reliability and productivity, and each configuration will subsequently be evaluated. This study seeks to provide a detailed understanding of possible operational improvements and their impact on the overall performance of the plant.

5.1. Stockpile Configuration Results

The incorporation of the stockpile in the plant was undertaken as a possible configuration aimed at improving system operational continuity. The implementation of the stockpile is intended to minimize the problem of low equipment utilization caused by the mutual influence between the two procedures [45]. The implementation of the stockpile configuration could provide a significant improvement in material availability and operational productivity, ensuring process continuity through the incorporation of intermediate storage systems. Figure 4a presents the initial configuration, where the stockpile system was not used. In contrast, Figure 4b shows the configuration with a stockpile system (highlighted by the green box), where a material storage and feeding system was implemented.

In

Table 1. Initial report.

Product	Capacity (t/h)	Product Percent (%)	Undersize (%)	Oversize (%)	Max Size (mm)	Gravel Percentage (%)
10	27	8	-	0	5	0
14	195	56	0	8	38	0
15	124	36	-	13	18	0
16	71	100	-	-	25	0

, it can be observed that the production capacity of some equipment is limited. For example, Product 14 has a capacity of 195 t/h, while Product 15 reaches only 124 t/h. After implementing the stockpile, production capacities significantly improved. Product 16 increased its capacity to 431 t/h, while Product 17 reached a capacity of 258 t/h (See

Table 2. Stockpile report.

Product	Capacity (t/h)	Product Percent (%)	Undersize (%)	Oversize (%)	Max Size (mm)	Gravel Percentage (%)
10	27	100	-	-	5	0
14	431	63	2	4	38	0
15	258	37	-	3	18	0
16	71	100	-	-	25	0

).

With the incorporation of the stockpile system, the production capacities of several products significantly increased. This change reflects greater efficiency in the material feeding process, resulting in higher total system productivity. The most notable improvement was in Product 16, whose operating capacity increased from 195 t/h to 431 t/h. Similarly, Product 15 increased from 124 t/h to 258 t/h, doubling its capacity.

The implementation of the stockpile reduced the criticality of the feeding equipment, as the intermediate storage system ensures the process continues without interruptions due to material shortages. This change not only increased system reliability but also improved the overall availability of the equipment, ensuring a constant production flow.

The stockpile configuration proved to be an effective strategy to improve reliability and operational productivity in comminution processes. The reduction in downtime, along with the increase in operating capacity, allows us to conclude that this configuration is suitable for ensuring the continuity of the production process and maintaining a constant flow of materials in operation.

5.2. Insertion of Equipment in Parallel Configuration

The implementation of the parallel configuration was carried out with the objective of optimizing production capacity by distributing the operational load among multiple pieces of equipment. This configuration is considered a potential improvement that could enhance both the reliability and productivity of the system as a whole. Figure 5 shows the incorporation of the parallel equipment, highlighted by the green box.

This configuration was tested to observe how the system behaved under different circumstances, ensuring that reliability and productivity did not decrease in the process.

Table 1 shows the initial report, where the capacity of Product 14 is 195 t/h.

Table 3. Parallel report.

Product	Capacity (t/h)	Product Percent (%)	Undersize (%)	Oversize (%)	Max Size (mm)	Gravel Percentage (%)
10	27	8	-	-	5	0
14	196	57	0	8	38	0
15	124	36	-	13	18	0
16	71	100	-	-	25	0

reflects the results obtained after the implementation of the parallel configuration, where the capacity of Product 14 increased to 196 t/h. Although this change is small, it indicates an improvement.

The parallel configuration allowed the operating load to be distributed between two crushers, which increased the production of Product 14. The “HP100” crusher is connected in parallel with another one of the same. The other products maintained a stable capacity, showing that the parallel configuration did not negatively affect its performance.

5.3. Fractional Configuration Insertion

The fractional configuration was applied to divide the operational load of critical equipment into several smaller pieces of equipment, with the goal of improving system reliability without compromising productive capacity. In Figure 6, the fractional configuration is depicted (highlighted by the green box). The operation of the critical HP300 equipment was divided between two HP200 and HP200e units, distributing the operational load. The results obtained are detailed below:

Product 14 has a capacity of 195 t/h. However, after implementing the fractionating configuration, as shown in

Table 4. Fractionating report.

Product	Capacity (t/h)	Product Percent (%)	Undersize (%)	Oversize (%)	Max Size (mm)	Gravel Percentage (%)
10	31	9	-	-	5	0
14	205	57	0	17	38	0
15	124	35	-	11	18	0
16	71	100	-	-	25	0

, the capacity of Product 14 increased to 205 t/h. Product 10 also showed improvement, increasing from 27 t/h to 31 t/h. Products 15 and 16 maintained their operational capacity without significant changes, indicating that the new configuration did not negatively affect their productivity. By dividing the load among several smaller pieces of equipment, the criticality of the original equipment was significantly reduced. This configuration is particularly useful in stages where the criticality of a single piece of equipment negatively affects system performance.

5.4. Monte Carlo Simulations

Monte Carlo simulation was used to evaluate and quantify whether the system reliability satisfied the operation and runtime requirement of the plant [46]. The parameters used in the simulation included data collected in an actual comminution site over three years, covering corrective maintenance of a mechanical, electrical, and instrumental nature. Additionally, two hundred random simulations were performed covering a one-year period.

To enhance the robustness of the simulation results, 95% confidence intervals were included to better reflect the dispersion and statistical reliability of the outcomes. Moreover, it is important to note that the RAM analysis is based on real reliability data. Prior to the simulation of new configurations, the probability distributions assigned to MTBF and MTTR values were validated within the APM platform to ensure their adherence to historical data, thereby reinforcing the representativeness and consistency of the simulation inputs. In this case, three Monte Carlo simulations were performed: the first with the initial configuration (Figure 7), the second with the parallel configuration (Figure 8), and finally with the fractional configuration (Figure 9).

The following results (

Table 5. Initial configuration histogram values. RMES.

Attribute	Percentage (%)
Mean	96.183
Median	96.288
Mode	96.281

) were obtained from the simulation in Figure 8:

The following results (

Table 6. Parallel configuration histogram values. RMES.

Attribute	Percentage (%)
Mean	96.889
Median	97.020
Mode	97.119

) were obtained from the simulation in Figure 9:

The following results (

Table 7. Fractionating configuration histogram values. RMES.

Attribute	Percentage (%)
Mean	96.312
Median	96.339
Mode	96.363

) were obtained from the simulation in Figure 9:

6. Discussion

The results obtained in this study allow for a critical evaluation of the analyzed configurations in terms of reliability and productivity. Based on the data obtained from the Monte Carlo simulation, it can be stated that the proposed configurations successfully increased the system’s reliability. The study’s primary innovation resides in the design and implementation of an integrated decision-support framework that bridges APM and AIP domains via a middleware algorithm. This middleware coordinates iterative simulations, enabling the evaluation of equipment configurations in terms of reliability indicators such as MTBF, MTTR, and criticality, alongside productivity levels and energy constraints. By moving beyond the siloed application of APM and AIP systems, the proposed methodology supports joint optimization across tactical and strategic layers, ensuring that operational realities and investment decisions are coherently aligned. Furthermore, the framework was applied in both early-stage design and mature operational phases, highlighting its flexibility and potential for broad application across various industrial sectors. This integrative approach represents a novel contribution to computational asset management and supports more informed, resilient, and sustainable configuration planning. The main findings for each configuration are discussed below.

The parallel configuration proved to be an effective option for improving system reliability by distributing the load among several pieces of equipment. This configuration reduced the criticality of key equipment and improved operational redundancy, allowing the system to maintain production even if one piece of equipment failed. The results showed that, although the increase in operational capacity was modest in some cases, the overall system reliability significantly increased.

The implementation of a fractional configuration led to a significant improvement in reliability by distributing the operational load of the most critical equipment among several smaller pieces of equipment. This allowed the system’s productivity to be maintained, with notable increases in some key products, such as Product 14. However, a limitation of fractionating is that the number of additional pieces of equipment can increase the operational complexity and energy consumption of the system. This must be carefully evaluated to ensure that the improvements in reliability are not counterbalanced by significant increases in operational complexity and energy consumption.

The stockpile configuration proved to be effective in improving material availability, which reduced downtime and allowed for more continuous operation. This approach is particularly useful in processes where material feed is intermittent or limited, as the stockpile acts as a buffer, ensuring a constant flow of materials.

A detailed quantitative analysis of the case study illustrates the tangible benefits achieved through the implementation of the proposed configuration strategies—stockpile insertion, parallel configuration, and fractionating or partial redundancy configuration. Each configuration was evaluated based on its impact on three core dimensions: productivity (measured as production capacity in t/h), reliability (assessed through availability histograms derived from Monte Carlo simulations), and energy efficiency (approximated via implied equipment usage and operational complexity).

In the baseline configuration, Product 14 exhibited a production capacity of 195 t/h, and Product 15 reached 124 t/h. The availability analysis using Monte Carlo simulation showed a mean availability of 96.183%, with a mode of 96.281%. Following the implementation of the stockpile configuration, Product 14’s capacity increased to 431 t/h (a 121% increase), and Product 15’s capacity rose to 258 t/h (a 108% increase). This sharp rise in throughput is attributed to the decoupling effect introduced by intermediate storage, which minimizes downtime due to upstream delays. Additionally, the stockpile configuration significantly enhanced operational continuity, though explicit availability metrics were not recalculated for this scenario. Given that no new active energy-consuming equipment was added, the energy efficiency of the system likely improved, as downtime and idling energy losses were reduced.

For the parallel configuration, Product 14’s output marginally increased from 195 t/h to 196 t/h, reflecting a 0.5% gain in productivity. However, the main advantage of this strategy lies in its impact on system reliability. The Monte Carlo simulation revealed an increase in mean availability to 96.889% and a mode of 97.119%, representing a +0.706-percentage-point gain in mean availability and a +0.838-point increase in mode availability over the baseline. This improvement highlights the value of load-sharing across redundant assets, which reduces criticality and enhances fault tolerance. Energy efficiency in this case must be evaluated more cautiously, as running two crushers in parallel may increase total energy consumption unless managed with appropriate load-balancing control strategies.

In the fractionating configuration, the load originally handled by a single HP300 unit was distributed across smaller HP200 and HP200e units. This resulted in an increase in Product 14’s throughput to 205 t/h, a 5.1% improvement over the original setup. Product 10 also saw a modest increase from 27 t/h to 31 t/h (approximately 15%). More importantly, the Monte Carlo simulation showed a mean availability of 96.312% and a mode of 96.363%, slightly higher than the baseline but lower than the parallel configuration. This suggests that while fractionation improves resilience by diluting criticality, it introduces complexity that may not always lead to dramatic reliability gains. Furthermore, the addition of more equipment units likely raises energy consumption, which must be balanced against reliability gains and operational throughput.

From the point of view of managerial insights, this work has shown that operational reliability can be significantly improved through the correct selection of configurations. Configurations such as the parallel and fractionating setups provide a favorable balance between reliability and productivity. However, each configuration has its limitations, and the choice of the best configuration largely depends on the specific characteristics of the mining operation.

From a practical standpoint, while the proposed methodology demonstrates clear improvements in reliability and productivity through configuration adjustments such as parallelization, redundancies, and the inclusion of stockpiles, its implementation in real industrial settings is subject to several operational constraints that must be carefully evaluated. One major limitation is physical space; for instance, implementing parallel or fractional configurations typically requires additional floor area to accommodate more equipment, conveyors, or auxiliary systems, which may not be available in existing facilities without significant retrofitting. Additionally, the constraints of existing infrastructure—including fixed foundations, pre-installed power distribution systems, and legacy control architectures—can significantly limit the feasibility of introducing new equipment or rearranging existing layouts. Even when space is available, integrating new configurations may necessitate lengthy downtimes or interruption of production, introducing opportunity costs that must be weighed against the expected performance gains. Moreover, capital investment requirements for acquiring and commissioning new equipment can be substantial, especially for high-capacity crushers or specialized feeders, and such investments must compete with other organizational priorities within constrained CAPEX budgets. There is also the issue of technical compatibility, as the integration of newer, more efficient equipment with older systems may require custom interfaces, retrofits, or additional training for operational staff. These practical considerations highlight the importance of conducting detailed feasibility studies, including physical layout analysis, cost–benefit evaluation, and change management planning, before transitioning from simulation-based recommendations to actual implementation. Thus, while the proposed optimization framework provides a powerful tool for identifying ideal configurations from a technical and operational perspective, its real-world deployment must be carefully adapted to site-specific conditions and constraints.

While the developed methodology offers a promising framework for integrating APM and AIP platforms to support equipment configuration decisions, certain limitations remain that suggest opportunities for further enhancement. One of the main areas for improvement lies in the economic dimension of the analysis. Although productivity, reliability, and energy efficiency are quantitatively assessed, the methodology does not yet include a detailed financial evaluation, such as investment costs, payback periods, or life cycle cost comparisons. Including these metrics would provide a more comprehensive foundation for management-level decision-making, especially in capital-intensive industries where financial justification is critical.

Additionally, while the case study effectively illustrates the methodology’s application in a comminution plant, broader industrial validation is still needed to confirm its scalability and adaptability. The current approach, including the use of a custom middleware and simulation tools, demonstrates technical feasibility, but its deployment in other sectors or under different operational constraints—such as limited physical space, integration with existing infrastructure, or equipment availability—has not yet been tested. Moreover, the model validation could be strengthened through more advanced statistical measures and sensitivity analyses to better reflect variability and uncertainty in real-world settings. These limitations do not undermine the value of the approach but rather highlight natural next steps for extending its robustness and practical relevance.

Future research could focus on the analysis of operational costs, an aspect that is emerging as fundamental for the development of this line of research. Delving into the maintenance costs and energy consumption of the selected configurations would allow for the identification of optimization opportunities that increase operational efficiency and profitability. The incorporation of environmental impact, from the perspective of costs associated, would offer an innovative approach to developing configurations aligned with current standards of environmental responsibility and carbon footprint reduction.

7. Conclusions

The integration of Asset Performance Management (APM) and Asset Investment Planning (AIP) represents a significant advancement in physical asset management. Emerging technologies have elevated the role of APM from a primarily operational monitoring tool to a strategic asset capable of informing complex investment decisions. This transformation is supported by a significant offer of data, advanced visualization, IoT technologies, and probabilistic analysis techniques.

Recent literature confirms the need for the integration of APM and AIP as a means of obtaining optimal asset configurations that enhance both reliability and productivity, without compromising energy efficiency or sustainability. This integrated approach offers a valuable contribution to the ongoing evolution of maintenance engineering, enabling data-driven strategies that align with long-term operational and sustainability goals. The proposed methodology for integrating APM and AIP platforms can leverage the joint optimization of productivity and reliability. Through simulation experiments, one can effectively optimize reliability and productivity in industrial projects.

To demonstrate the validity of the proposed integration framework, a case study was presented. Throughout this study, various system configurations (parallel, fractionation, and stockpiling) were evaluated, identifying those that reduced the criticality of key equipment and improved operational efficiency. Alternative configurations can be tested, and based on the results and the specific needs of the user, the most appropriate configuration can be selected to maximize both the reliability and productivity of the system.

The integration of technical data from the equipment and reliability criteria has allowed for the selection of configurations that maintain or increase productivity without compromising operational continuity. Among the evaluated configurations, the parallel and fractionating setups stood out as the most effective for improving system reliability. The use of Monte Carlo simulations validated that the proposed configurations increase the system’s reliability. In particular, the fractionation alternative showed a significant improvement in reliability, raising the modal value from 96.281% to 97.435%.

The findings obtained provide a solid foundation for alternative and more efficient and sustainable implementations of more suitable configurations of a production system. By identifying and optimizing equipment configurations at an early stage, operational parameters are established that can be adapted and applied in real environments. This anticipation allows for a smoother transition to large-scale operations, minimizing risks and ensuring greater efficiency from the start and for the long term.

The methodology proposed for integrating Asset Performance Management (APM) and Asset Investment Planning (AIP) platforms, while developed and validated within the context of the mining industry, holds significant potential for application across a variety of other asset-intensive industries where equipment reliability, operational continuity, and long-term investment decisions are closely intertwined. One prominent example is the energy and utilities sector, particularly in power generation and water treatment facilities, where asset life cycles are long, equipment failures can have severe regulatory and service delivery implications, and capital investment decisions are highly scrutinized. In these contexts, the integration of real-time condition monitoring data from APM systems with the long-term planning models of AIP platforms can support proactive decisions regarding asset replacement, system redundancy, or reconfiguration, ultimately enhancing reliability and reducing total life cycle costs.

Another sector where this approach could be highly impactful is manufacturing, especially in process industries such as chemical production, pulp and paper, or food and beverage, where production throughput and energy efficiency are critical performance metrics. In these environments, the ability to simulate alternative configurations, evaluate their impact on operational KPIs, and incorporate reliability data into investment planning can drive both productivity improvements and cost savings. Similarly, in transportation and logistics infrastructure—including rail systems, ports, and airport facilities—the coordinated integration of APM and AIP can help manage complex networks of interdependent assets, optimizing maintenance schedules and capital upgrades in a way that minimizes service disruption and aligns with long-term capacity planning. Across these sectors, the methodology’s use of simulation, reliability analysis, and energy performance evaluation provides a structured and scalable decision-support tool for aligning operational realities with strategic asset investment goals.

The proposed integration framework constitutes a solid foundation for the optimization of configurations in production systems. Although there are opportunities for improvement, the obtained results are promising and provide a robust starting point for future research and practical applications. This approach allows progress towards more reliable, efficient, and sustainable productive operations. Therefore, future research should explore the economic implications of the proposed configuration changes by incorporating detailed life cycle cost (LCC) models and cost–benefit analyses. This would allow for a more comprehensive assessment of the trade-offs between improved reliability, increased productivity, and the capital and operational expenditures associated with different asset configurations. Additionally, integrating financial indicators into the decision-making framework would enhance its practical value, particularly in capital-intensive industries where investment decisions must be justified through measurable returns.

Another future research direction involves the extension of the current methodology to incorporate environmental performance indicators, such as carbon footprint, water usage, or energy intensity metrics. This would align the framework with sustainability goals and emerging regulatory requirements. Moreover, developing automated processes for the integration of AIP and APM platforms—potentially through middleware solutions, APIs, or AI-driven data processing—could streamline configuration evaluations, reduce human error, and enable real-time optimization. These advancements would significantly enhance the scalability and adaptability of the proposed methodology across diverse industrial contexts.