Enhancing Efficiency in Custom Furniture Production with Intelligent Scheduling Systems

Abstract

With the upgrading of consumption driving the transformation of the home furnishing industry towards personalized customization, panel furniture enterprises are confronted with a core contradiction between large-scale production and individualized demands: The traditional production management model is unable to cope with the chaos in production scheduling, resource waste, and low collaborative efficiency caused by small-batch and multi-variety orders. This paper proposes an intelligent production scheduling system that integrates Enterprise Resource Planning (ERP), Manufacturing Execution System (MES), Advanced Planning and Scheduling (APS), and Warehouse Management System (WMS), and elaborates on its data processing methods and specific application processes in each production stage. Compared with the traditional model, it effectively overcomes limitations such as coarse-grained planning, delayed execution, and information islands in middle-level systems, achieving deep collaboration between planning, workshop execution, and warehouse logistics. Empirical studies show that this system not only can effectively reduce the production costs of customized panel furniture manufacturers, enhance their market competitiveness, but also provides a digital transformation framework for the entire customized panel furniture manufacturing industry, with significant theoretical and practical value.

1. Introduction

In recent years, the increasingly prominent pursuit of life quality by consumers has made “customized furniture” a hot topic [1,2]. The customized furniture market provides consumers with a higher added value of life, enabling them to design and adjust the home environment according to their personal needs. Many customized furniture enterprises seized the market opportunity and achieved rapid expansion of the company’s scale [3]. As a result, the furniture industry has shifted from the traditional seller-led and semi-industrialized production model to a buyer-oriented production model of multiple varieties and small batches [4]. Driven by Industry 4.0, the current large-scale customized production model has gradually formed.

The characteristic of the panel furniture industry is to produce products composed of multiple panel pieces according to the requirements of customers. However, the large number of individualized orders and small batch sizes make the production planning and scheduling very difficult. In the production process, it faces various challenges in the production (complex non-standard panels), information (data integration barriers), and management (dynamic coordination inefficiencies) [5].

In view of the complexity and uncertainty in the process of furniture manufacturing and production, more scientific control and scheduling methods are needed to ensure the efficiency and smoothness of furniture production [6]. As early as the 1960s to 1980s, scholars proposed to guide enterprises to achieve high-quality and low-loss production through MRP (Material Requirement Planning), JIT (Just in time) production, and the Theory of Constraints [7,8], and proposed the introduction of information technology to support the scheduling optimization of the production process [9]. In the early stage of modern management, most enterprises adopted ERP (Enterprise Resource Planning) systems for information management. However, with technological progress and the expansion of production scale, Vanessa et al. [10,11] found that the application of MES (Manufacturing execution system) can more effectively track the production process and conduct real-time scheduling, providing support for the development of Industry 4.0.

The concept of planned production scheduling was first seen in process manufacturing [12,13]. With the continuous development of industrial management, this concept has gradually been introduced to other manufacturing fields, such as the furniture production industry. Chen [14] developed a production management system for furniture enterprises with MRP-II as the core, improving the level of production management. Xiong and Wu [15] analyzed the differences between custom furniture and traditional batch production, emphasizing that large-scale customization relies on information technology. Bao et al. [16] proposed transforming the traditional furniture manufacturing workshop through the MES to achieve efficient digital production scheduling. Some scholars [17,18] further developed an integrated production scheduling system by combining ERP and MES to meet the needs of personalized and large-scale production scheduling.

The existing MES and ERP systems have four limitations in the application of the furniture manufacturing industry: first, the dynamic scheduling mechanism is lacking, making it difficult to respond promptly to the volatility and insertion demands of multi-variety and small-batch orders [19]; second, the collaboration between systems is broken. The information isolation of ERP/MES leads to cross-departmental data barriers and the centralized architecture is difficult to handle complex order flows [20]; third, real-time monitoring is weak and the status of equipment and materials cannot be tracked at the second level [21]; fourth, intelligent decision-making is insufficient, lacking predictive analysis, and visual support [22]. These defects expose that the traditional production scheduling architecture has not yet achieved the vertical integration from equipment to management, the horizontal collaboration of the supply chain, and the digital connection of the entire process, constituting the main obstacles to customized transformation.

To address these limitations, this paper proposes an integrated intelligent production scheduling system that combines ERP, MES, APS, and WMS. The motivation behind this work is to overcome the disconnection and inefficiencies in existing systems and to respond dynamically to the demands of mass customization in panel furniture manufacturing. The key research questions include: (1) How can these systems be effectively integrated? (2) How should they be applied in real production settings to support small-batch, large-scale manufacturing of panel furniture? These questions guide the development, implementation, and evaluation of the proposed system in this study.

2. Production Scheduling Management System of Panel Customized Furniture Company

2.1. Service Architecture and Practical Implementation of the Production Scheduling Management System

In the early stages of manufacturing management system development, traditional production systems commonly adopted an ERP-MES architecture. In this structure, the ERP system is responsible for enterprise-level resource planning and production order release, while the MES handles shop-floor operations and execution data collection. However, this architecture lacks the ability to perform detailed production planning and optimization. The master production plans generated by ERP are often too coarse-grained to accurately match shop-floor resources and real-time conditions, leading to resource conflicts, bottleneck accumulation, and ultimately, reduced manufacturing efficiency [23,24].

To address these limitations, many enterprises have introduced Advanced Planning and Scheduling (APS) systems [25], positioned between ERP planning and MES execution, to provide fine-grained scheduling and optimized allocation of production resources and tasks.

Despite this improvement, system integration often remains limited to pairwise collaborations [26]—such as ERP-MES or ERP-APS—which, while achieving partial data connectivity, lack direct communication between core mid-layer systems. As a result, the response loop between production scheduling and shop-floor execution is extended, creating significant information gaps. On one hand, APS must rely on ERP as an intermediary to access MES feedback, preventing it from dynamically adjusting schedules based on real-time production data and thus reducing scheduling effectiveness. On the other hand, real-time shop-floor anomalies collected by MES [27] (e.g., bottleneck delays, equipment failures) cannot be directly fed back to the APS system, resulting in deviations between planned and actual production and lowering the overall integration efficiency and data consistency. This loosely coupled mid-layer and tightly coupled top-down structure is increasingly inadequate for supporting modern, responsive, and flexible manufacturing environments [28].

Against this backdrop, the custom furniture industry, in an effort to improve the traditional management system structure (Figure 1a), has proposed an improved system architecture—ERP-3S-PCS (as shown in Figure 1b). The “3S” system comprises APS, MES, and WMS, forming an integrated mid-layer that retains the traditional layered control logic while significantly enhancing horizontal connectivity among modules. This eliminates data silos between systems and enables seamless linkage among scheduling, execution, and logistics, thereby improving both scheduling responsiveness and shop-floor execution efficiency [29].

Compared with traditional three-tier architectures, the ERP-3S-PCS model offers a more clearly defined modular structure and stronger system synergy. It is better suited to the demands of customized production, such as rapid order insertion [30], dynamic resource reallocation, and cross-functional coordination—demonstrating greater flexibility, scalability, and responsiveness in modern manufacturing settings [15].

2.2. Planning Layer in the Production Scheduling Management System

As the core management platform of an enterprise, the ERP system has six core functions: engineering data management, production management, project management, customer service management [31], material management, and financial management [32]. It can coordinate the work of various departments, optimize the allocation of enterprise resources, improve management efficiency, and be market-oriented to fully support various operational activities of the enterprise.

2.3. Executive Layer in the Production Scheduling Management System

The executive layer is divided into three parts: WMS, MES, and APS [33,34,35]. WMS is responsible for inventory management, goods tracking, and inbound and outbound management, and monitors the information and status of warehouse goods in real time [36]. APS optimizes the production sequence based on constraints such as the production calendar and process routes, and allocates workshops and production lines according to product attributes to ensure capacity matching. MES focuses on workshop production process management, covering personnel management, material management, equipment management, production tracking, and production statistics. It tracks the position of workpieces and the progress of tasks through means such as RFID and PDA, records production data in real time, and generates visual reports to assist management in decision-making. The specific functions it covers are shown in Figure 2.

The optimized production scheduling management system enhances the flexibility and response speed of the production plan through the ERP-3S-PCS structure, adapting to customized demands and market changes. In the specific implementation of the production scheduling management system, the APS, MES, and WMS systems are responsible for production plan optimization, workshop execution management, and warehouse material management, respectively [37]. Through data collection, analysis, and resource allocation, it ensures the efficient operation and quality control of the production process. Figure 3 shows the specific implementation method of this process.

2.4. APS—MES Collaborative Mechanism

In the ERP-3S-PCS architecture, a closed-loop control mechanism is established between APS and MES to support real-time synchronization between scheduling and execution, in alignment with the ISA-95 enterprise-control system integration standard [38]. APS, based on order requirements, bills of materials (BOM), process routes, and resource constraints, employs advanced mathematical models and algorithms—drawing from Theory of Constraints (TOC), Hybrid Flow Shop Scheduling (HFSP), and batch optimization theory—to generate optimal production schedules, which are then delivered to MES. MES organizes shop-floor operations accordingly and continuously collects key execution data in real time, including output, task progress, equipment status, and quality metrics. In the event of anomalies—such as production delays, equipment failures, or material shortages—MES immediately feeds back the information to APS. The APS system then dynamically adjusts the schedule by rescheduling tasks, reallocating resources, or revising material supply plans, thereby correcting deviations and preventing further disruptions [24].

This mechanism forms a continuous loop of “planning–execution–feedback–re-planning,” [39] enabling iterative optimization and real-time data-driven scheduling. MES not only serves as the executor of plans but also acts as a real-time sensing terminal for shop-floor conditions; APS, on the other hand, continuously refines scheduling strategies based on incoming feedback. This bidirectional data exchange enables highly flexible and adaptive scheduling, ensuring that production plans remain aligned with actual shop-floor conditions and effectively support the demands of customized and flexible manufacturing environments [16].

2.5. Advantages and Generalizability of the Architecture

Compared to the traditional ERP-MES structure, the ERP-3S-PCS architecture introduces APS, MES, and WMS as an integrated mid-layer, enabling seamless interaction among planning, execution, and logistics modules. This design replaces the loosely connected, pairwise system collaborations with a tightly coupled, real-time closed-loop. APS generates fine-grained, dynamically optimized schedules based on real-time feedback from MES, which monitors production progress, resource status, and anomalies on the shop floor. WMS, in turn, ensures timely material handling and quality control through synchronized warehouse operations. This coordinated system improves scheduling accuracy, shortens response times to disruptions, and enhances overall production flexibility—particularly beneficial for customized, high-mix, low-volume manufacturing like panel-based furniture [40].

Beyond the furniture industry, this architecture has strong applicability in other discrete manufacturing sectors such as electronics assembly, home appliances, and precision machining—where rapid order handling, high variability, and complex process coordination are critical. However, its effectiveness hinges on the enterprise’s automation level and digital maturity, including reliable equipment connectivity, standardized system interfaces, and real-time data collection infrastructure [41]. For small-scale manufacturers with stable, low-complexity orders, the return on investment in advanced modules like APS may be limited. Therefore, the deployment of ERP-3S-PCS should be tailored to operational needs, with careful cost-benefit evaluation and a phased, modular rollout strategy [42].

3. Data Processing in Production Scheduling Management

The production process of panel furniture is heavily reliant on data support. From the system’s front-end to the production back-end, every stage necessitates substantial data input through Industrial Internet of Things technologies. A data-driven production model not only enhances production efficiency but also guarantees product uniformity and high quality. These data permeate multiple phases, including design, production, warehousing, and shipping, optimizing and monitoring the entire production process [36,43]. The subsequent sections will provide a detailed elaboration on the classification, collection methods, flow processes of the data, and its impact on production decisions.

3.1. Data Category

In the production process of custom panel furniture, the primary data types involved encompass material data, production data, equipment data, and personnel data. To develop effective plans and organize production efficiently, it is essential to comprehensively and accurately acquire this information [36,43].

(1)

Order Data. When a customer places an order, the order enters the design phase, where the designer completes the design work. Following this, the Middleware platform reviews the design to ensure compliance with standards. Upon successful review, the system automatically generates the material list required for the product using the material-unwrapping software. At the production facility, the planner can access details such as the customer’s expected delivery date, delivery city, and production base code, enabling the formulation of a detailed production plan.

(2)

Material Data. The material data encompasses various dimensions, primarily comprising material basic coding information, material circulation records, inventory levels, and material quality inspection results. Specifically, the material basic coding information includes, but is not limited to, panel coding, color coding, and edge banding coding; the material circulation records document the timing of material entry and exit, quantities, and their application scenarios; the inventory levels detail the usage status and stock balance of artificial panels, edge banding strips, veneer materials, and hardware components. In the context of production scheduling management informatization for panel custom furniture companies, accurately capturing warehouse inventory data is of paramount significance. The material control department must verify the material quantities according to order specifications and submit the production material procurement plan to the procurement department at least five working days in advance to ensure the seamless and efficient operation of production processes.

(3)

Production Data. The production data encompass essential information spanning the entire production process, from raw material cutting to the final product packaging and warehousing. This includes, but is not limited to, the reporting nodes for each process and detailed records of any abnormalities during panel processing. Specifically, the reporting nodes document the start and completion times of panels at each stage, as well as their circulation status between different processes. The abnormal information meticulously logs various issues encountered during operations such as cutting, edge banding, and drilling [44].

(4)

Equipment Data. The equipment data information encompasses details such as equipment codes, fault records, tool scheduling information, and inspection and maintenance logs. In production scheduling, the precise allocation of specific production volumes and schedules to individual pieces of equipment is essential. During the panel manufacturing process, any equipment failure in operations such as material cutting, edge banding, and drilling must be promptly reported to enable timely adjustments to the production line. Following the completion of maintenance and repair, a thorough evaluation of the equipment’s performance will be conducted, and a report based on these findings will be submitted to the planning department to inform the development of subsequent production plans [45].

3.2. Workshop Data Collection Methods

In the production process of customized panel furniture components, to achieve effective production tracking and a comprehensive understanding of the overall production status in the workshop. To accurately track the panel components during production and integrate detailed processing information, a label is assigned to each panel immediately after it is cut by the electronic panel saw. The content structure of this label is illustrated in Figure 4.

The label specifies the panel’s order number, process route code, punching position, color pattern, thickness, and edge banding dimensions, among other details. Each QR code label serves as the electronic identity certificate for the panel. During subsequent processing, operators need only scan the QR code to retrieve processing instructions from the human-computer interaction interface, ensuring the accuracy of the process. Data collection primarily utilizes bar code scanners and handheld terminal devices, such as Personal Digital Assistants (PDAs) [46].

In the component sorting stage, prior to entering the warehouse, specialized scanning equipment will scan the labels. Upon completion of the scanning process, the system will allocate storage locations and loads based on the current storage status of the stereoscopic warehouse and the order numbers. Once the orders are finalized, the robotic arm will sort the components to the logistics area to await shipment.

3.3. Data Flow Procedure

The data flow process vividly illustrates the production process of panel-customized furniture. Upon receiving orders from the company’s direct stores or dealers, designers undertake product design and modeling according to customer customization requirements. These orders are subsequently forwarded to the order center for processing, including confirming delivery dates and generating BOM, drawings, and task orders. Once the design plan and quotation have been approved, the system transmits detailed order information to the 3S system to execute tasks such as order splitting, optimization, material procurement, and scheduling. Subsequently, the MES dispatches execution commands to the Process Control System (PCS), which provides feedback on the execution outcomes. Concurrently, MES must continuously collect real-time process parameters and equipment performance data.

The PCS or the supervisory control and data acquisition system (SCADA) provides essential data, including equipment numbers, equipment statuses, operating parameters, task volumes, finished product quantities, and production process information, to the MES.

3.4. Data Analysis and Visualization

During the production process, a variety of reports and charts are generated through data tracking. These data are visually presented on multiple terminals, including electronic reports, workshop KANBAN (Figure 5) [47,48], and tablet computers, enabling easy access to key information such as the completion rate of panel orders, the number of orders entering the warehouse, the occupancy status of inventory locations, and the volume of orders awaiting processing. This visual representation provides robust decision-making support for both workshop managers and the corporate decision-making level [49].

3.5. Feedback Effect of Production Data on Production Scheduling

The production data from the workshop plays a crucial role in providing feedback for enterprise production scheduling [49]. For instance, by examining the daily production capacity data of a specific production line, one can intuitively grasp the processing volume, the remaining batches for the day, and the estimated time required to process the remaining panels based on the production rhythm, all displayed on the workshop data dashboard via the SCADA [50]. This information enables a judgment on whether the processing tasks will be completed by the predetermined production deadline. Should there be an expectation of delay, the workshop supervisor must promptly inform the planning engineer, review the data from other production lines, assess their available production capacity, and reallocate some orders to these lines to ensure task completion [51,52].

4. Instance Verification

4.1. Research Methodology

The purpose of this case study is as follows: (1) Select the leading panel furniture customization company in the industry to fully demonstrate the entire production process from customer customization to the production of the product and its transportation to the customer. Through the operation process of the production management system architecture described in the previous text, improve efficiency, reduce production time, and the delivery period. (2) Verify the advantages of the batch production mode. Select the wardrobe production line, observe the production process of the panel pieces in the production workshop: cutting, edge banding, drilling, and formulate a mathematical model formula for production time. Write a scheduling program with infinite cache constraints using MATLAB, select 80, 200, and 2000 panels for production scheduling experiments, and compare the processing time through Gantt charts to verify the advantages of the batch production strategy in large-scale panel production. Figure 6 presents the flowchart of the steps used in the research process.

4.2. A Comprehensive and Systematic Description of the Production Process

This case is sourced from a leading customized home furnishing enterprise. The enterprise has a mature and standardized production process in the field of panel customized furniture and is highly representative. The production process is as follows:

Figure 7a illustrates the centralized submission of regional customer orders to the middleware system for material decomposition. First receiving orders from customers in various regions, the company generates a unique sales order code for each order. These orders are then uniformly submitted to the company’s middleware platform system for material decomposition processing. Different material decomposition methods can significantly influence the time required for production data to be transmitted from the front end to the manufacturing back end. The primary material decomposition software utilized within the company includes the Zowee V2.0 and DSC V1.0 [53].

Figure 7b shows the routing of decomposed orders to different factories based on order source and capacity, ensuring efficient task allocation. Upon completion and approval of the material decomposition of the order, the relevant data will be transmitted to the backend manufacturing facility for production scheduling. The data primarily include the sales order number, anticipated processing completion time, panel piece name, panel piece color code, and dimensions, among other details. Prior to formulating the production plan, it is essential to ascertain the specific production location for the task order based on the order source, order attributes, and production capacity ratios of each factory, specifying that the corresponding panel pieces are manufactured on designated production lines.

Figure 7c presents the rolling N + 5 scheduling mechanism that enhances production-sales alignment and emergency responsiveness. After the order allocation is completed, it enters the production scheduling stage, which is handled by each planning engineer. Each planner is responsible for the production scheduling of one or more production lines. During the scheduling process, production orders are arranged in accordance with the N + 5 rolling production plan, i.e., the production calendar, to improve the accuracy of the plan, ensure the consistency of production and sales, and effectively deal with emergencies such as urgent orders, supplementary orders, and goods shortages.

Figure 7d demonstrates the order batch grouping strategy based on panel similarity, improving scheduling efficiency. When the planners are doing the order scheduling, in order to improve the production efficiency, they group different orders into the same batch according to the similarities of the panel to be produced [54], such as the color pattern, the glue used, the size, the thickness, etc., and will make minor adjustments according to the actual production needs to ensure that the panel produced in each batch are within the appropriate range.

Figure 7e shows the panel optimization and cutting path generation process, which ensures high material utilization and efficient instruction delivery. After the batch grouping optimization is completed, further optimization steps will be carried out. The main purpose of this step is to ensure that, in the cutting process, as many panels as possible can be cut from a large sheet of material, thereby improving the utilization rate of the sheet. After the optimization, specific cutting paths will be generated and sent to the processing equipment. At the same time, to facilitate the cutting operation, similar cutting paths are usually generated for multiple sheets in a batch grouping for laminated panel production. If the utilization rate of the sheet in the generated optimization scheme is lower than 20%, the scheme will be abandoned and these panels will be cut and processed on the remaining materials from the previous links in the production workshop. If the size of the panel to be produced exceeds the maximum size of the sheet material library (usually 2440 × 1220), adjustments to the modeling scheme are required [55,56].

After the optimization process is completed, the system will generate cutting paths for all batches of panels on the production line within a single day and automatically produce the material requisition form. This form specifies the exact specifications and quantities of large panels required for production, thereby facilitating the collection of materials from the warehouse management department by production workshop personnel. Furthermore, the system will also generate either paper or electronic production task orders to provide detailed production instructions to operators at each workstation.

Figure 7f outlines the predictive material ordering and verification process to ensure timely and accurate supply to production lines. The warehouse purchasing department utilizes historical material demand data and a predictive model to forecast material requirements and place orders in advance at predetermined intervals. Concurrently, following the N + 5 optimization strategy outlined by the planning department [57], it verifies the completeness of materials, ensuring that production requisites are delivered to the production line both timely and accurately, thereby guaranteeing the seamless execution of the production schedule.

Figure 7g details the four-stage production process, with real-time data-driven execution. When the workshop production line receives the required materials, the panel production operation can commence. The manufacturing of panels for customized furniture typically involves four stages: cutting, edge banding, drilling, and packaging. During the cutting stage, the equipment automatically retrieves the cutting diagram to process the panel and completes label posting via the HMI and a label printer. The HMI provides operational instructions for label application. The primary responsibility of the main operator is to apply labels and transport the panels to the buffer area. In subsequent processes such as edge banding and drilling, the processing equipment is equipped with a scanner gun or PDA device to scan the production labels on the panels, thereby obtaining specific processing techniques and methods. Different processing techniques correspond to distinct process codes and paths, and the appropriate processing programs are dynamically invoked in real-time during production. In the packaging process, reflective infrared sensors are installed adjacent to the conveyor belt to precisely measure the dimensions of the panels and subsequently cut the corresponding packaging shells [58]. During this process, multiple panels are stacked according to the principle of placing larger panels at the bottom and smaller ones on top.

Figure 7h depicts the final logistics handover stage. Once all panels for a given order have been fully produced and assembled into sets, the logistics department takes over to load the vehicle and initiate the transportation process, ensuring timely delivery to customers.

Adopting the aforementioned production mode can effectively streamline and integrate the workflows of various departments, thereby enhancing support for workshop operations. Through batch-optimized production planning, not only is the utilization rate of sheet materials significantly improved, but also similar panel parts are grouped for production, substantially reducing equipment setup time and increasing equipment operational efficiency. Furthermore, the close collaboration between the purchasing and planning departments provides robust planning and material assurance, further boosting workshop production capacity [59]. As illustrated in Figure 8, the enhanced equipment operational efficiency notably contributes to increased production capacity.

4.3. Verification of Batch Grouping Optimization Strategy

The wardrobe cabinet body production line of this company is selected as the experimental object, as customized wardrobes are typically composed of multiple regular rectangular panels, which allows for the standardized calculation of processing times and facilitates subsequent optimization research. The layout of the production line equipment is illustrated in Figure 8. Specifically, there are two sets of equipment dedicated to material cutting and edge sealing processes, respectively, and three sets of equipment for the drilling process. In Figure 9, m_1,1 denotes the first set of equipment in the first process, while m_2,2 indicates the second set of equipment in the second process, and so on.

In this study, 80 and 200 panels were selected for verification. These panels encompass various types, including laminates, side panels, top panels, bottom panels, and back panels, all of which constitute the wardrobe cabinet body structure.

Through on-site inspection of the workshop and consideration of actual production conditions, the processing time for each panel at each stage was meticulously calculated based on the dimensions and structural types of the components [60].

(i)

Measurement Methodology for Determining the Processing Time in the Cutting Process

In panel furniture production, large panels are typically cut into smaller ones using two-dimensional rectangular nesting. The process involves first cutting the panel into long strips along its short edge (Figure 10a), then rotating the strips 90° to cut them into smaller panels along the length (Figure 10b).

Based on the panel sawing method described above, let a_j (mm) and b_j (mm) represent the long and short sides of the panel, respectively. The total travel distance of the saw blade, denoted as S_k,j (mm), follows a reciprocating motion and can be calculated using Equation (1).

$$S_{k,j}=2\times 2\left(a_j+b_j\right)$$

(1)

Since the processing time T_k,j can be expressed as the ratio of the cutting distance S_k,j to the equipment feed rate U_k, and given that the feed rate is consistent across all equipment in the same process, the processing time T_k,j can be formulated as Equation (2):

$$T_{k,j}={\displaystyle \frac{1}{\alpha }}\left({\displaystyle \frac{3S_{k,j}}{50U_k}}\right)$$

(2)

α denotes the fatigue allowance coefficient during the cutting process60, utilized for empirically adjusting the ideal processing time, with a value of 0.523. k signifies the sequence number of the current process, which is 1.

(ii)

Measurement Methodology for Determining the Processing Time in the Edge Banding Process

The four-line edge banding machine is the most widely utilized edge banding equipment in this enterprise. This machine enables the panel to undergo a single-pass, four-sided edge banding process via conveyor belt circulation. Not only does it effectively protect the core material, but it also enhances the aesthetic appeal of the finished product. Figure 11 shows the edge banding path of a panel.

The panel is first conveyed a distance d₁ to the edge banding machine, where the first long side a_j is processed across a worktable of length d₂. It then moves through angled rollers for alignment and bands the second long side. Afterward, the panel rotates 90° via rollers of length d₃ to begin banding the two short sides b_j. The total travel path S_k,j during this process is defined by Equation (3).

$$S_{k,j}=2\times \left(d_1+d_2\right)+d_3+a_j$$

(3)

Thus, the processing time T_k,j of the panel satisfies Equation (4):

$$T_{k,j}={\displaystyle \frac{3S_{k,j}}{50\beta \times U_k}}$$

(4)

β represents the fatigue allowance coefficient of the edge sealing process, which is used for empirical correction of the ideal processing time and its value is 0.523. k represents the current edge sealing process and is 2.

(iii)

Measurement Methodology for Determining the Processing Time in the Drilling Process

The drilling processing time for the panel is typically determined by the distribution and number of holes. Common hole position distributions, as illustrated in Figure 12, can be categorized into three types: (i) exclusively horizontal holes; (ii) exclusively vertical holes; (iii) a combination of horizontal and vertical holes. The arrows in the figure indicate the direction of the processing holes on the panel. All hole positions adhere to the specifications of the 32 mm system, enabling the row drill to achieve the processing effect of an entire group of holes in a single operation, thus ensuring efficient and rapid processing.

Based on the distribution characteristics of various hole positions, the drilling processing time for the panel can be categorized and analyzed systematically.

(i)

Exclusively Horizontal Holes

The row drill equipment, equipped with two spindles for simultaneous hole processing on both sides of the workpiece, adopts a consistent method for estimating processing time, equally applicable to scenarios requiring horizontal hole processing on only one side or both sides. The spindle motion trajectory is: initially translating from its starting position to the front of the target hole; then performing an axial feed to complete drilling; and finally returning to the starting position. Consequently, the moving distance S_k,j for drilling a horizontal hole is always equal to the hole depth D_horizontal plus the distance H1 from the drill bit’s initial position to the workpiece surface. Given process requirements, both D_horizontal and H₁ are fixed values, and the entire drilling operation can be characterized as a reciprocating motion, with S_k,j calculated by Equation (5).

$$S_{k,j}=2\times \left(D_{horizontal}+H_1\right)$$

(5)

(ii)

Exclusively Vertical Holes

The processing procedures for vertical holes are largely analogous to those for horizontal holes, with the primary distinction being the depth of the vertical holes D_vertical, and the distance from the initial position of the drill bit to the workpiece surface H₂. The feed distance during vertical hole drilling can be represented by Equation (6):

$$S_{k,j}=2\times \left(D_{vertical}+H_2\right)$$

(6)

(iii)

A Combination of Vertical and Horizontal Holes

Next, we analyze the scenario involving the mixed processing of horizontal and vertical holes. The motion trajectory and distance are consistent with those described earlier. However, horizontal and vertical holes cannot be processed concurrently. Typically, all horizontal holes must be completed before processing any vertical holes. Consequently, in this case, the movement distance S_k,j of the drill bit can be expressed by Equation (7):

$$S_{k,j}=2\times \left(D_{vertical}+D_{horizontal}+H_1+H_2\right)$$

(7)

Through the analysis of the above three types of drilling, it can be finally concluded that the processing time T_k,j of each panel in the drilling process satisfies Equation (8):

$$T_{k,j}={\displaystyle \frac{3S_{k,j}}{50\gamma \times U_k}}$$

(8)

γ represents the fatigue allowance coefficient of the edge sealing process, while K represents the coefficient value of the current drilling process, and its value is 2.

According to

Table 1. Scheduled Production Orders for a Specific Production Line.

Order Number	Panel Type	Color	Size
A2024080Y1	Door panel	Hibiscus pattern	400 × 300 × 18
YY24100756	Laminated panel	Silver pear pattern	2045 × 650 × 25
YY24100352	Tatami panel	Calico pattern	482 × 552 × 18
L20240906Y6	Laminated panel	Calico pattern	899 × 380 × 25
A20240912Y7	Laminated panel	Lychee pattern	923 × 279 × 25
A20240912Y6	Laminated panel	Dark-elm pattern	881 × 300 × 25
202409205Y36	Side panel	Kwanzaa oak	820 × 303 × 18
…
F20140907Y13	Back panel	Kwanzaa oak	463 × 703 × 18
The total number of panels: 80

, the initial production plan comprised 80 panels with varying patterns and sizes. To minimize the manufacturing cycle and reduce equipment setup changes, the orders were optimized by consolidating similar items. As illustrated in

Table 2. Order Batches for Batch Optimization Production.

Batch Number	Number of Panels	Product Line
241031-1001	5	A001-1
241031-1002	10	A001-1
241031-1003	8	A001-1
241031-1004	12	A001-1
241031-1005	7	A001-1
241031-1006	9	A001-2
241031-1007	6	A001-2
241031-1008	11	A001-1
241031-1009	8	A001-2
241031-1010	4	A001-2

, the 80 panels were reorganized into 10 batches. Specifically, panels with identical or similar patterns were grouped into the same batch and allocated to different production lines based on their respective capacities to ensure efficient processing.

(iv)

Analysis of the Dispatching Results

By estimating the processing time of the panels in each process, an actual production scheduling simulation comparison is conducted.

In the aforementioned production process analysis, equipment adjustments are typically required for processing panel components with varying attributes. Specifically, the setup times for material cutting, edge banding, and drilling processes are 20 s, 45 s, and 10 s, respectively. Figure 13 illustrates the scheduling outcomes after implementing the batch grouping strategy, the different colors in the picture represent different batches of components.. Compared to the non-batch grouping approach, this strategy significantly reduces equipment adjustment time, thereby markedly shortening the overall processing time. Moreover, as the number of panel components increases, the disparity in completion time between the batch grouping and non-batch grouping scheduling methods becomes more pronounced. Figure 13a shows that the manufacturing completion time for the non-batch grouping method is 1318 s. However, when the products are grouped into 10 main batches based on the color mode, the completion time is reduced to 1278 s, as shown in Figure 13b. Similarly, in Figure 13c, the manufacturing completion time for 200 ungrouped panels is 3208 s, while grouping them into 13 batches based on the color mode reduces the completion time to 2866 s (Figure 13d). In the production scheduling simulation of 2000 panels, Figure 13e,f show that the minimization of the maximum completion time has decreased from 36,210 s without batching to 25,675 s after batch production, achieving a 30.2% optimization of the production cycle. These results highlight the advantages of the batch grouping strategy in large-scale production and its significant role in improving production efficiency.

5. Conclusions

Based on the current situation of the panel custom furniture market, this paper analyzes the need of such enterprises for a detailed and perfect production management system, systematically introduces the functions and positioning of the current production scheduling management system, and elaborates in detail, combined with the actual production process.

• By integrating market dynamics with specific business needs, the enterprise’s production scheduling management system has established a three-tier architecture comprising ERP, 3S (APS + MES + WMS), and PCS. This structure facilitates seamless data interoperability and enhances communication and collaboration across departments, thereby significantly boosting production efficiency.
• In the production process, given that processing batches are divided into the main line and auxiliary line, it is essential to meticulously plan the processing volume and production lead time for each batch in order to achieve balanced operation of the production line.
• During the production scheduling process, order coding rules serve as an effective foundation for grouping and production guidance. To enhance interdepartmental collaboration efficiency, it is necessary to further unify the coding methods and panel classification criteria across different systems.
• During the production process, rework panels currently depend on manual error reporting and handling for transportation. There is an absence of a fully automated system for error correction, inspection, and reverse logistics tracking.
• The 3S system proposed in this study offers valuable guidance for small and medium-sized enterprises (SMEs) undergoing transformation toward Industry 4.0. This architecture helps lower the barriers to adopting intelligent manufacturing and enhances overall operational efficiency.
• Certain limitations may arise during practical implementation. Weak digital infrastructure or limited technical personnel, system integration, and employee training can pose significant challenges. Moreover, the relatively small order volumes and high variability in production schedules typical of many SMEs may limit the effectiveness of APS in large-scale scheduling optimization, potentially leading to underutilization of system resources.
• Future research could further explore the applicability of this architecture to other product types, particularly in fields characterized by high-mix, low-volume production. Additionally, integrating Artificial Intelligence (AI) and Machine Learning (ML) technologies could enhance the system’s intelligence in areas such as scheduling prediction, anomaly detection, and real-time decision-making.