A batch household appliance maintenance cycle determination method and system, electronic equipment and storage medium

Abstract

The application discloses a batch household appliance maintenance period determination method, comprising the following steps: determining a wear correction maintenance period according to batch common characteristics of current household appliances of the current type, daily average use time of the current household appliances and household appliance historical fault information; determining a service adaptation factor according to platform service data and the wear correction maintenance period; determining a service correction maintenance period according to the service adaptation factor and the wear correction maintenance period; determining a sudden failure probability according to a reference maintenance period, the service correction maintenance period and a preset basic failure risk threshold; judging whether the sudden failure probability is greater than the preset basic failure risk threshold; if yes, adjusting a target maintenance period threshold. The application quantifies the correlation between common characteristics and failure risk, calculates an optimal maintenance period, improves the maintenance efficiency of enterprises and reduces the maintenance cost. The application also discloses a system, an electronic device and a storage medium for implementing the above method.

Description

Technical Field

[0001] This invention relates to the field of computer data processing technology, and in particular to a method, system, electronic device, and storage medium for determining the maintenance cycle of batch home appliances. Background Technology

[0002] As enterprises accelerate their digital transformation, bulk home appliances (typically ranging from dozens to hundreds of units) in scenarios such as hotels and office buildings have become core operational assets. For enterprise users with large-scale homogeneous home appliance assets, such as hotels, chain apartments, commercial office buildings, and large supermarkets, the stable operation of home appliances is a core element in ensuring service quality. However, currently, enterprises generally face a dilemma when managing bulk home appliances (such as hundreds of air conditioners in hotels and water dispenser clusters in office buildings): "insufficient maintenance leading to frequent downtime" and "excessive maintenance leading to soaring costs."

[0003] For the maintenance of such appliances, existing solutions often adopt fixed maintenance cycles (such as "maintaining every six months"). To be on the safe side, managers often tend to shorten the maintenance cycle (such as changing it from 180 days to 90 days), which leads to excessive consumption of spare parts, inflated labor costs, and frequent disassembly may introduce the risk of human-caused damage. If the cycle is set too long, frequent sudden failures will result in high emergency repair costs and business downtime losses. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention provides a method, system, electronic device, and storage medium for determining the maintenance cycle of batch home appliances. The technical problem to be solved by this invention is achieved through the following technical solution: The first aspect of this invention provides a method for determining the maintenance cycle of batch home appliances, comprising the following steps: The batch common feature coefficients are determined based on the batch common features of multiple current home appliances of the current type; among which, the batch common features include: brand consistency feature, model consistency feature, installation time difference feature, usage scenario consistency feature, and average daily usage time difference feature; Determine the baseline maintenance cycle based on historical fault information of the current type of home appliance; The batch loss coordination coefficient is determined based on the batch common characteristic coefficient and the average daily usage time of the current home appliances; The loss correction maintenance cycle is determined based on the baseline maintenance cycle and the batch loss coordination coefficient. The service matching factor is determined based on the number of available service personnel on the platform, the number of home appliances maintained by a single service personnel in a single visit, the current total number of home appliances, and the wear and tear correction maintenance cycle. The service correction and maintenance cycle is determined based on the service adaptation factor and the loss correction and maintenance cycle. The probability of sudden failure is determined based on the baseline maintenance cycle, the service correction maintenance cycle, the total number of home appliances, and the preset basic fault risk threshold. Determine whether the probability of the sudden failure is greater than the preset basic failure risk threshold; If so, the target maintenance cycle is calculated iteratively based on the Poisson distribution and preset adjustment coefficients until the probability of multiple home appliances failing is less than the preset demand risk threshold.

[0005] In one embodiment of the present invention, the batch common feature coefficients The calculation formula is: in, This indicates brand consistency. Indicates model consistency characteristics. Indicates the installation time difference characteristic. This indicates consistency in usage scenarios. This indicates the characteristic of the difference in average daily usage time. These represent the corresponding preset weight coefficients.

[0006] In one embodiment of the present invention, the historical fault information of home appliances includes: the number of historical fault home appliance samples of home appliances whose similarity to the batch common features is greater than or equal to a preset threshold, and the number of samples of home appliances with historical faults. The time of failure of a historical home appliance sample and the first Average daily usage time of a sample of historical home appliances; The benchmark maintenance cycle The calculation formula is: in, M This represents the number of historical fault samples of home appliances whose similarity to the common features of the batch is greater than or equal to a preset threshold. Indicates the first The time of failure for a sample of historical home appliances; Indicates the first Average daily usage time of a sample of historical home appliances; This represents the fail-safe factor.

[0007] In one embodiment of the present invention, the batch loss coordination coefficient The calculation formula is: in, This indicates the average daily usage time of the current home appliance, in hours.

[0008] In one embodiment of the present invention, the correction maintenance cycle The calculation formula is: .

[0009] In one embodiment of the present invention, the service adaptation factor The calculation formula is: in, This indicates the number of available service personnel on the platform. This indicates the number of home appliances maintained by a single service personnel in a single visit. This indicates the total number of home appliances currently in use.

[0010] In one embodiment of the present invention, the service correction and maintenance cycle The calculation formula is: .

[0011] A second aspect of the present invention provides a system for determining the maintenance cycle of batch home appliances, comprising: The first determining module is used to determine the batch common feature coefficients based on the batch common features of multiple current home appliances of the current type; wherein, the batch common features include: brand consistency feature, model consistency feature, installation time difference feature, usage scenario consistency feature, and average daily usage time difference feature; The second determining module is used to determine the baseline maintenance cycle based on historical fault information of the current type of home appliance. The third determining module is used to determine the batch loss coordination coefficient based on the batch common characteristic coefficient and the average daily usage time of the current home appliances; The first correction module is used to determine the loss correction maintenance cycle based on the baseline maintenance cycle and the batch loss coordination coefficient. The fourth determination module is used to determine the service adaptation factor based on the number of available service personnel on the platform, the number of home appliances maintained by a single service personnel in a single visit, the current total number of home appliances, and the wear and tear correction maintenance cycle. The second correction module is used to determine the service correction and maintenance cycle based on the service adaptation factor and the loss correction and maintenance cycle. The calculation module is used to determine the probability of sudden failure based on the baseline maintenance cycle, the service correction maintenance cycle, the total number of home appliances, and the preset basic fault risk threshold. The judgment module is used to determine whether the probability of the sudden failure is greater than the preset basic failure risk threshold. The iteration module is used to iteratively calculate the target maintenance cycle based on the Poisson distribution and preset adjustment coefficients if the probability of multiple home appliances failing is less than the preset demand risk threshold.

[0012] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for determining the maintenance cycle of batch home appliances provided in the first aspect of the present invention.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method for determining the maintenance cycle of batch home appliances provided in the first aspect of the present invention.

[0014] The beneficial effects of this invention are: This invention avoids repetitive calculations for thousands of individual devices by quantifying and aggregating the common characteristics of a batch of home appliances, significantly reducing data processing volume and computing power consumption. Through a batch loss synergy coefficient, it quantifies the synergistic loss effect of multiple appliances within the same physical space. Dynamic correction of this coefficient ensures that the calculation results are not merely a collection of theoretical values, but truly reflect the physical loss patterns of the equipment under actual operating conditions, effectively solving the maintenance cycle deviation problem caused by traditional algorithms neglecting the "group effect." Furthermore, through a batch maintenance service adaptation factor, it ensures that the generated maintenance cycle service instructions are truly executable. Finally, by calculating the correlation with fault risk, it improves the enterprise's maintenance efficiency and reduces maintenance costs.

[0015] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 A flowchart illustrating a method for determining the maintenance cycle of bulk home appliances according to an embodiment of the present invention; Figure 2 This is a block diagram of a system for determining the maintenance cycle of bulk home appliances, provided in an embodiment of the present invention. Detailed Implementation

[0018] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0019] like Figure 1 As shown, the first aspect of this invention provides a method for determining the maintenance cycle of batch home appliances, comprising the following steps: Step 11: Determine the batch common feature coefficients based on the batch common features of multiple current home appliances of the current type.

[0020] Among them, the common characteristics of batches include: brand consistency, model consistency, installation time difference, usage scenario consistency, and average daily usage time difference.

[0021] Step 12: Determine the baseline maintenance cycle based on the historical fault information of the current type of home appliance.

[0022] Step 13: Determine the batch loss coordination coefficient based on the batch common characteristic coefficient and the average daily usage time of the current home appliances.

[0023] Step 14: Determine the loss correction maintenance cycle based on the baseline maintenance cycle and the batch loss coordination coefficient.

[0024] Step 15: Determine the service matching factor based on the number of available service personnel on the platform, the number of appliances maintained by a single service personnel in a single visit, the current total number of appliances, and the wear and tear correction maintenance cycle.

[0025] Step 16: Determine the service correction and maintenance cycle based on the service adaptation factor and the loss correction and maintenance cycle.

[0026] Step 17: Determine the probability of sudden failure based on the baseline maintenance cycle, service correction maintenance cycle, total number of home appliances, and preset basic fault risk threshold.

[0027] Step 18: Determine whether the probability of a sudden failure is greater than the preset basic failure risk threshold.

[0028] Step 19: If so, the target maintenance cycle is iteratively calculated based on the Poisson distribution and the preset adjustment coefficient until the probability of multiple home appliances failing is less than the preset demand risk threshold.

[0029] In this embodiment, by quantifying and aggregating the common features of a batch of home appliances, repetitive calculations for thousands of individual devices are avoided, greatly reducing data processing volume and computing power consumption. The batch loss synergy coefficient quantifies the synergistic loss effect of multiple home appliances within the same physical space. Through dynamic correction of this coefficient, the calculation results are no longer merely a collection of theoretical values, but truly reflect the physical loss patterns of the equipment under actual operating conditions, effectively solving the maintenance cycle deviation problem caused by traditional algorithms neglecting the "group effect." The batch maintenance service adaptation factor ensures that the generated maintenance cycle service instructions are truly executable. Finally, by calculating the optimal maintenance cycle based on the correlation with fault risk, the enterprise's maintenance efficiency is improved and maintenance costs are reduced.

[0030] Based on the first aspect of the present invention, the second aspect of the present invention provides a method for determining the maintenance cycle of batch home appliances in further detail. The second aspect of the present invention provides a method for determining the maintenance cycle of batch home appliances, applied to a service platform, and includes the following steps: Step 201: Determine the batch common feature coefficients based on the batch common features of multiple current home appliances of the current type.

[0031] Among them, the common characteristics of batches include: brand consistency, model consistency, installation time difference, usage scenario consistency, and average daily usage time difference.

[0032] Prior to this step, enterprise users have already entered basic information about a batch of home appliances into the service platform used in this embodiment, forming a ledger that includes the quantity of home appliances. Types of home appliances (such as air conditioners, washing machines), brands ,model Installation time Average daily usage time .

[0033] The service platform has accumulated historical fault data for similar home appliances (same brand, same model, same scenario): fault occurrence time Fault type, maintenance duration, and maintenance cost.

[0034] The service platform has entered the following parameters for on-site service resources: the number of home appliances that a single service personnel can maintain in a single visit. Duration of a single home visit Number of service personnel.

[0035] In this step, for a batch of home appliances of a certain type from a company, common features of the batch are extracted to obtain a set of common features of the batch. : , Indicates brand consistency (values ​​0-1, 1 indicates all are the same brand). This indicates model consistency (value 0-1, 1 indicates all are the same model). This indicates the installation time difference characteristic (in a batch of home appliances, the earliest 5% and the latest 5% of the appliances are removed, and the latest installation time is taken as the difference between the earliest installation time and the latest installation time (unit: days). For example, in 50 air conditioners, if the first and last 5 are removed, the earliest installation is the 1st and the latest installation is the 8th, and the installation time difference is 7 days). This indicates the consistency of usage scenarios (values ​​range from 0 to 1, with 1 indicating the same usage scenario). Indicates the difference in average daily usage time ( ).

[0036] Batch common feature coefficients The calculation formula is: Preset weighting coefficients: It can be adjusted according to the type of home appliances owned by the business, and preset. , , The closer it is to 1, the more obvious the common characteristics of a batch of home appliances are, and the more suitable it is for unified planning of maintenance cycles.

[0037] In this step, the brand ( ) and model ( ) is the core influencing factor of the wear and tear pattern of home appliances, therefore it has the highest weight (0.25 each); installation time difference ( ) and the difference in average daily usage time ( The larger the value, the worse the batch consistency, therefore... Quantization of decreasing function ( The larger the value, the smaller the value of this item (avoiding distortion of coefficients due to excessively large single differences); consistency in usage scenarios ( The environmental impact on appliance wear and tear is secondary (0.15). It is a slowly increasing function. When C≥0, the value range is (0,1], which differs from other features ( The range of values ​​for the comprehensive coefficient is consistent to ensure that the comprehensive coefficient is within the same range. The dimensions are unified, so it can be directly used for subsequent calculations.

[0038] In this step, five common characteristics of mass-produced home appliances ( Quantized into "batch common feature coefficients" of a unified dimension This is used for subsequent correction and maintenance cycles, and its core is to reflect that "the more consistent the batch, the higher the feature coefficient, and the more uniformly the maintenance cycle can be planned."

[0039] Example: bulk home appliancesN =50 units, all from Gree ( ), the same model KFR-35GW ( ), installation time difference All rooms have air conditioning (days / nights). ), Daily usage time difference Hours (max8.5h - min8h = 0.5); Weight ; Calculation process: calculate Corresponding item: ; calculate Corresponding item: ; = 0.25×1 + 0.25×1 + 0.2×0.325 + 0.15×1 + 0.15×0.712 ≈ 0.25+0.25+0.065+0.15+0.107 ≈ 0.822; The value is close to 1, indicating that the batch of air conditioners has obvious common characteristics and is suitable for unified planning of maintenance cycles, without the need to formulate individual plans for each unit.

[0040] Step 202: Determine the baseline maintenance cycle based on the historical fault information of the current type of home appliance.

[0041] In this step, based on historical fault data of similar home appliances with common characteristics in a batch, the theoretical maintenance cycle is calculated when there is no batch-coordinated loss, providing a benchmark for subsequent corrections. The core is to ensure that the benchmark cycle conforms to the batch pattern based on historical data of the batch.

[0042] Historical appliance malfunction information includes: the number of historical malfunctioning appliances that have a similarity to a preset threshold of common features in a batch. M , No. Failure time of a historical home appliance sample and the Average daily usage time of a sample of historical home appliances .

[0043] Standard maintenance cycle The calculation formula is: Among them, the platform has accumulated a large number of historical faulty home appliances of the same type that have a "similarity of ≥80% with common features to the current batch". M ≥30, to ensure sample validity; The unit is days, the total number of days from the date of installation completion to the time of the malfunction; Unit: hour; This represents the fault safety factor (1.05-1.15, generally preset to 1.1, used to reserve safety redundancy and avoid unreasonable baseline periods due to historical sample bias). In this step, The unit is days, which represents the theoretical maintenance cycle required to avoid malfunctions in a batch of similar home appliances without any associated wear and tear.

[0044] In this step, only samples with "similarity ≥ 80% to common features of the current batch" are selected (similarity = current batch). The ratio of the common characteristic coefficients of historical sample batches ensures that the wear and tear patterns of historical data are consistent with those of the current batch of home appliances. As a weighting factor, since usage time is a core influencing factor on appliance wear and tear (the longer the usage time, the earlier the failure occurs), a weighted average can avoid the influence of single-sample bias on the baseline period. Considering the sensitivity of enterprise users to "losses due to sudden failures," β is introduced for conservative correction to ensure that the baseline period has a certain safety redundancy, while controlling... The baseline period should be kept between 1.05 and 1.15 to avoid exceeding the historical failure time and ensure reasonableness.

[0045] For example, the hotel currently has 50 Gree air conditioners, and the platform has accumulated a large number of historical faulty home appliance samples with a similarity of ≥80%. M =50 units, The core data of the historical samples (taking 10 representative samples, with the remaining 40 samples calculated using the same logic) are shown in Table 1: Table 1 Calculation process: Weighted average time to failure: ; Standard maintenance cycle sky; The baseline maintenance cycle for this batch of air conditioners is approximately 179 days. Historical failure times are concentrated between 155 and 165 days. The baseline cycle is only about 17 days longer than the weighted average failure time, allowing for reasonable safety redundancy and avoiding excessively long cycles that could increase the risk of failure, thus ensuring rationality.

[0046] Step 203: Determine the batch loss coordination coefficient based on the batch common characteristic coefficient and the average daily usage time of the current home appliances.

[0047] In this step, the impact of "mass use of home appliances" on wear and tear is quantified (e.g., simultaneous start-up and shutdown of hotel air conditioners will exacerbate compressor wear and lead to premature failure; concentrated use of printers in office buildings will result in poor heat dissipation and increased wear and tear). The mass wear and tear coordination coefficient is used to correct the baseline maintenance cycle, making the cycle more consistent with the actual mass use scenario.

[0048] Batch loss coordination coefficient The calculation formula is: in, This represents the average daily usage time of the current home appliance, in hours. Constraints: ∈[0.8,1.2] (Boundary constraints are used to avoid abnormal synergy coefficients; this range is verified based on a large amount of batch appliance wear data).

[0049] here, The smaller the value (the worse the batch common characteristics), the closer λ is to 1.2 (the more aggressive the collaborative loss). The larger the value (the more frequently it is used), the closer λ is to 1.2 (the more intense the collaborative loss), and vice versa, the closer λ is to 0.8 (the more intense the collaborative loss).

[0050] In this step, This reflects the principle that "the worse the batch consistency, the more complex the collaborative losses"—a batch of home appliances comes from mixed brands and models, with significant differences in installation time and inconsistent wear patterns during use. Collaborative effects exacerbate overall losses, therefore λ increases with... It increases as it grows. The term reflects the principle that "the higher the usage frequency, the more significant the collaborative loss"—the higher the average daily usage time and the higher the frequency of simultaneous operation of home appliances, the more severe the collaborative loss. Therefore, λ increases with... The weights 0.2 and 0.1 are calibrated based on a large amount of batch home appliance failure data (verified through linear regression, these weights can most accurately quantify the impact of the two factors on the synergistic loss).

[0051] Step 204: Determine the loss correction maintenance cycle based on the baseline maintenance cycle and the batch loss coordination coefficient.

[0052] Adjust maintenance cycle The calculation formula is: .

[0053] here, When (cooperative losses intensify), (Shorten maintenance cycles and prevent premature failures); When (cooperative losses are mitigated), (Extend maintenance cycles and reduce maintenance costs).

[0054] Example, =0.822, Hours (average daily usage time of 50 air conditioners). =179 days. Calculation process: Batch loss coordination coefficient = 1 + 0.2×(1-0.822) + 0.1×(8.25 / 10) ≈ 1.1181; Maintenance cycle after loss co-correction = 179 / 1.1181 ≈ 160.1 days ≈ 160 days.

[0055] here, This indicates that the coordinated losses of this batch of air conditioners have intensified (because although the batch is relatively consistent, the average daily usage time is high, and the coordinated losses from start-up and shutdown are obvious). Therefore, the baseline period of 179 days needs to be shortened to 160 days to better reflect the actual loss patterns and avoid premature failures.

[0056] Step 205: Determine the service matching factor based on the number of available service personnel on the platform, the number of appliances maintained by a single service personnel in a single visit, the current total number of appliances, and the wear and tear correction maintenance cycle.

[0057] In this step, the maintenance cycle is adjusted based on the service capabilities of the on-site service platform to ensure the feasibility of the batch maintenance plan and solve the disconnect problem of "only calculating the maintenance cycle without considering service capabilities".

[0058] Service Adaptation Factor The calculation formula is: in, This indicates the number of available service personnel on the platform (the number of service personnel that the platform can allocate). This indicates the number of appliances maintained by a single service personnel in a single visit (default). k =4 units / person / time (can be adjusted according to the type of home appliance). This indicates the total number of home appliances currently in use. 30 indicates that each month is calculated based on 30 days (a unified time dimension). This represents the number of months corresponding to the maintenance cycle after loss compensation (rounded up to ensure sufficient service cycles). Constraints: μ ∈(0,1], μ =1 indicates that the platform's service capabilities are fully adapted to batch maintenance needs; μ <1 indicates insufficient service capacity, requiring a shorter maintenance cycle (by splitting service batches to reduce the service pressure on each batch).

[0059] Here, molecules This represents the maximum number of batch maintenance services the platform can provide per month (number of service personnel × maintenance volume per person × number of days per month); denominator This represents the total service volume required for a batch of home appliances within their maintenance cycle (total number of appliances × number of months corresponding to the maintenance cycle, i.e., "number of appliances requiring maintenance per month"); min avoids the adaptation factor exceeding 1 (when service capacity is excessive, there is no need for excessive adjustment; maintain a constant level). μ =1 is sufficient).

[0060] Step 206: Determine the service correction and maintenance cycle based on the service adaptation factor and the loss correction and maintenance cycle.

[0061] Service repair and maintenance cycle The calculation formula is: .

[0062] In this step, When (service capacity is insufficient), (Shorten maintenance cycles and split service batches). hour, (No corrections required).

[0063] Example, known: tower, sky, One service staff member Calculation process: per person per session The number of months corresponding to the maintenance cycle: Months; Maximum number of maintenance tasks per month for the platform: = 3 × 4 × 30 = 360 units; Number of home appliances requiring monthly maintenance: N = 50 × 6 = 300 units; Service Adaptation Factor μ = min(1, 360 / 300) = 1; Maintenance cycle after service adaptation correction = 160 × 1 = 160 days.

[0064] Description of insufficient service capacity: like One service staff member Taiwan / person / time; The maximum number of systems that can be maintained per month on the platform is 2 × 4 × 30 = 240. μ =min(1, 240 / 300)=0.8; At this point, the 50 air conditioners need to be split into two batches (25 units per batch), with one batch maintained every 128 days to ensure that the platform's service capacity can support them.

[0065] Step 207: Determine the probability of sudden failure based on the baseline maintenance cycle, service correction maintenance cycle, total number of home appliances, and preset basic fault risk threshold.

[0066] In this step, the maintenance cycle after correction is verified. Can the probability of sudden failures of mass-produced home appliances be controlled within an acceptable threshold for the company? If it exceeds the threshold, the cycle needs to be further shortened to ensure that the core objective of "reducing downtime losses due to sudden failures" is achieved.

[0067] Probability of sudden failure The calculation formula is: in, This represents the preset basic fault risk threshold, which is the basic probability of a "sudden failure of a single home appliance" that the company can tolerate (default 0.05, i.e., 5%). Specifically, it is the basic probability of a "sudden failure of a single home appliance" that the company can tolerate. This is the natural constant (approximately 2.718), used to fit the Poisson distribution, simplifying fault probability calculations and adapting to the platform's practical computing power requirements. Here... This represents the calculated probability that at least one appliance in a batch will experience a sudden malfunction (unit: decimal, e.g., 0.08 is 8%).

[0068] The core function of this formula is to "quantify the current maintenance cycle". Below, the overall risk of sudden failures in batches of home appliances—the larger the batch size (the larger N is), the longer the maintenance cycle ( The larger the value, the higher the overall failure probability; the baseline period The longer the length, the more acceptable the threshold for businesses. The smaller the value, the lower the overall failure probability. The form essentially uses a Poisson distribution to approximate the probability of batch failures (batch appliance failures can be regarded as random independent events, which conform to the characteristics of a Poisson distribution), avoiding complex product calculations, reducing the platform's computing power requirements, and ensuring that the calculation accuracy meets practical needs.

[0069] Step 208: Determine whether the probability of sudden failure is greater than the preset basic failure risk threshold. If it is less than or equal to the preset basic failure risk threshold, proceed to step 209. If it is greater, proceed to step 210.

[0070] Step 209, if : Explain the current situation Reasonable, the risk of failure is controllable, and no adjustment is needed.

[0071] Step 210, if This indicates the current situation. The length is too long, and the risk of failure exceeds the acceptable range for the enterprise, requiring iterative adjustments. .

[0072] In this step, the company's actual need is that the maximum number of appliances experiencing sudden malfunctions does not exceed a certain tolerance threshold. Therefore, the target maintenance cycle can be iteratively calculated based on a Poisson distribution and a preset adjustment coefficient until the probability of multiple appliances malfunctioning is less than the preset risk threshold. Since the tolerance for a single appliance malfunction is... In batch scenarios, to ensure overall risk control, the requirement is that "the number of failures does not exceed..." The probability of this event occurring is at least [percentage missing]. .

[0073] here, This represents the maximum number of home appliances required by the company to withstand sudden malfunctions. To preset the demand risk threshold, the calculation process based on the Poisson distribution is as follows: X follows a Poisson distribution. ,calculate If it is not true, then adjust according to the iterative formula. Iterative adjustment (for a preset adjustment coefficient, for example) The value is 0.9, indicating a 10% reduction each time. The calculation continues after adjustment. If the condition is not met, the iterated value is used to continue calculating the adjusted service correction and maintenance cycle in the adjustment formula until... Once established, the adjusted cycle at this point becomes the target maintenance cycle.

[0074] Example, known: =50 units, =160 days =179 days =0.05 (5%); Calculation process: Calculate the core parameters: ; Probability of sudden batch failures %); In the formula, R represents the physical meaning of "the probability of at least one appliance suddenly failing," while what businesses actually accept is "the number of failing appliances ≤ 5% of the total number." Therefore, the verification standard is as follows: In this example, X ≤ 50 × 5% = 2 appliances. Whether it is valid or not.

[0075] X follows a Poisson distribution. ≈2.2345; Poisson distribution probability formula: ; Calculation process: P(X=0) = ≈ 0.107 × 1 = 0.107; P(X=1) = ≈ 0.107 × 2.2345 ≈ 0.239; P(X=2) = ≈ 0.107 × 4.993 ≈ 0.264; It does not meet the verification standards and requires iterative adjustments. First iteration: = 0.9 × 160 = 144 days; = ; ≈ 0.134 × 5.033 ≈ 0.674 < 0.95; Second iteration: = 0.9 × 144 = 129.6 days ≈ 130 days; = ; ≈ 0.177 × 4.3088 ≈ 0.763 < 0.95; Third iteration: = 0.9 × 130 = 117 days; = ; ≈ 0.197 × 3.941 ≈ 0.776 < 0.95; Fourth iteration: = 0.9 × 117 = 105.3 days ≈ 105 days; = ; ≈ 0.234 × 3.5065 ≈ 0.821 < 0.95; Fifth iteration: = 0.9 × 10⁵ = 94.5 days ≈ 95 days; ; ≈ 0.268 × 3.1795 ≈ 0.852 < 0.95; Sixth iteration: = 0.9 × 95 = 85.5 days ≈ 86 days; ; P(X≤2) ≈ e^(-1.1955)×(1+1.1955+0.714) ≈ 0.302×2.9095 ≈ 0.879<0.95; Seventh iteration: = 0.9 × 86 = 77.4 days ≈ 77 days; ; ≈ 0.344 × 2.635 ≈ 0.907 < 0.95; Eighth iteration: = 0.9 × 77 = 69.3 days ≈ 69 days; ; ≈ 0.386 × 2.4055 ≈ 0.929 < 0.95; Ninth iteration: = 0.9 × 69 = 62.1 days ≈ 62 days; ; ≈ 0.420 × 2.241 ≈ 0.941 < 0.95; Tenth iteration: = 0.9 × 62 = 55.8 days ≈ 56 days; ; ≈ 0.457×2.088 ≈ 0.954 ≥ 0.95.

[0076] After calculations in steps 201-210, the optimal target maintenance cycle for a batch of home appliances is finally output. (After verification), and based on enterprise needs and platform service capabilities, a batch maintenance plan will be generated: Optimal maintenance cycle Specify the time interval for each maintenance batch (e.g., the final 56 days in the example); Maintenance batch division: If the platform's single-batch service capacity is insufficient, batches will be split. For example, 50 air conditioners will be split into 2 batches of 25 units each, with a maintenance interval of [missing information]. / Batch quantity; On-site service arrangements: Based on the service personnel and service duration, clarify the service time, service personnel, and service process for each batch; Fault warning: Based on algorithm-based fault risk calculation, the platform will remind you 7 days in advance to arrange on-site service and remind the company to prepare for cooperation 3 days in advance.

[0077] Furthermore, based on actual data (maintenance records, fault records) after a period of on-site maintenance, the preset weighting coefficients can be dynamically optimized. This allows maintenance cycles to increasingly align with the actual wear and tear patterns of a company's batch of home appliances, achieving a closed loop of "data-driven → precise planning".

[0078] In this step, specifically, the iterative optimization formula for the preset weight coefficients is (gradient descent method): in, j =1,2,3,4,5 (corresponding to the weights of the 5 common features in the batch) ), t This indicates the number of iterations (t≥1, one iteration is performed for every 100 batches of batch maintenance data). Indicates the first t After the nth iteration j A preset weighting coefficient, The learning rate (0.01-0.05, default 0.03, controls the iteration step size to avoid unstable convergence due to excessively fast iteration) and the loss function are also specified. (Quantify the deviation between the "target maintenance cycle" and the "actual optimal maintenance cycle"). Indicates the first Target maintenance cycle for batches of home appliances Indicates the first The actual optimal maintenance cycle for batches of home appliances (calculated based on actual fault data and maintenance costs, i.e., the cycle with the minimum "fault loss + maintenance cost"). K This indicates the total number of batches accumulated, typically 100 batches. Represents the loss function pair The partial derivatives (for simplified calculation, the "difference approximation" is used in practice): Where Δα = 0.01 (a small perturbation used to calculate the difference).

[0079] By continuously adjusting the weight coefficients using the gradient descent method , making the loss function L Minimize, that is, minimize the deviation between the "maintenance cycle calculated by the algorithm" and the "actual optimal maintenance cycle". Control the iteration step size. η An excessively large weight may cause the iteration to fail to converge (due to excessive fluctuations in the weight coefficients). η Too small a value might lead to slow iterations; 0.03 is a reasonable value validated based on platform service data. Loss function. LMean squared error (MSE) is used to quantify the magnitude of the deviation. MSE can amplify larger deviations and avoid "small deviations accumulating and causing algorithm failure". Differential approximation is used to replace complex partial derivative calculations, which reduces the platform's computing power requirements and facilitates practical implementation.

[0080] Example, known: current iteration number t =1, =0.25, η =0.03, Δα=0.01, K =100 batches of maintenance data, (Loss value) ; Calculation process: Approximate partial derivative: (13.5 - 12.8) / 0.01 = 70; The second iteration = 0.25 - 0.03×70 = 0.25 - 2.1 = -1.85 (Out of weight range, boundary constraints required); Boundary constraints: (Ensure each weight has a reasonable proportion), therefore =0.05; Verify the adjusted loss: (The reduction in losses indicates that the adjustment was effective.)

[0081] This invention focuses on the batch management and control needs of enterprise users for home appliances. By aggregating and analyzing common features, it solves the pain points of low efficiency and high cost of single home appliance maintenance algorithms, and is suitable for batch home appliances in multiple scenarios such as hotels and office buildings.

[0082] This invention boasts high accuracy by introducing a batch loss coordination coefficient and a service adaptation factor, combined with fault risk verification and iterative optimization, to ensure that the maintenance cycle conforms to the wear and tear patterns of home appliances while avoiding fault risks, resulting in an accuracy improvement of over 40% compared to existing algorithms.

[0083] This invention is deeply integrated with the on-site service platform scenario, incorporating service capabilities into algorithm calculations to avoid the problem of "reasonable maintenance cycle but inability to be implemented". It can directly connect with entrepreneurs' radio accounts and platform service resources to automatically generate executable maintenance plans.

[0084] This invention iteratively optimizes parameters using the gradient descent method. Based on the actual data accumulated on the platform, it can adapt to the wear differences of different brands and types of home appliances. The accuracy continues to improve after long-term use, without the need for manual intervention.

[0085] like Figure 2 As shown, a third aspect of the present invention provides a system for determining the maintenance cycle of batch home appliances, comprising: The first determining module 31 is used to determine the batch common feature coefficients based on the batch common features of multiple current home appliances of the current type; wherein, the batch common features include: brand consistency feature, model consistency feature, installation time difference feature, usage scenario consistency feature and daily average usage time difference feature; The second determining module 32 is used to determine the baseline maintenance cycle based on the historical fault information of the current type of home appliance. The third determining module 33 is used to determine the batch loss coordination coefficient based on the batch common characteristic coefficient and the average daily usage time of the current home appliances; The first correction module 34 is used to determine the loss correction maintenance cycle based on the baseline maintenance cycle and the batch loss coordination coefficient. The fourth module 35 is used to determine the service matching factor based on the number of available service personnel on the platform, the number of home appliances maintained by a single service personnel in a single visit, the current total number of home appliances, and the wear and tear correction maintenance cycle. The second correction module 36 is used to determine the service correction and maintenance cycle based on the service adaptation factor and the loss correction and maintenance cycle. Calculation module 37 is used to determine the probability of sudden failure based on the baseline maintenance cycle, service correction maintenance cycle, total number of home appliances and preset basic fault risk threshold; The judgment module 38 is used to determine whether the probability of a sudden failure is greater than a preset basic failure risk threshold. The iteration module 39 is used to iteratively calculate the target maintenance cycle based on the Poisson distribution and the preset adjustment coefficient if the probability of multiple home appliances failing is less than the preset demand risk threshold.

[0086] In one embodiment of the present invention, batch common feature coefficients The calculation formula is: in, This indicates brand consistency. Indicates model consistency characteristics. Indicates the installation time difference characteristic. This indicates consistency in usage scenarios. This indicates the characteristic of the difference in average daily usage time. These represent the corresponding preset weight coefficients.

[0087] In one embodiment of the present invention, the historical fault information of home appliances includes: the number of historical fault home appliance samples that have a similarity to a batch of common features greater than or equal to a preset threshold, and the number of samples with historical faults. The time of failure of a historical home appliance sample and the first Average daily usage time of a sample of historical home appliances; Standard maintenance cycle The calculation formula is: in, M This represents the number of historical fault samples of home appliances whose similarity to common features in the batch is greater than or equal to a preset threshold. Indicates the first The time of failure for a sample of historical home appliances; Indicates the first Average daily usage time of a sample of historical home appliances; This represents the fail-safe factor.

[0088] In one embodiment of the present invention, the batch loss coordination coefficient The calculation formula is: in, This indicates the average daily usage time of the current home appliance, in hours.

[0089] In one embodiment of the present invention, the maintenance cycle is modified. The calculation formula is: .

[0090] In one embodiment of the present invention, the service adaptation factor The calculation formula is: in, This indicates the number of available service personnel on the platform. This indicates the number of home appliances maintained by a single service personnel in a single visit. This indicates the total number of home appliances currently in use.

[0091] In one embodiment of the present invention, the service repair and maintenance cycle The calculation formula is: .

[0092] A fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method for determining the maintenance cycle of batch home appliances provided by the present invention described above.

[0093] A fifth aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method for determining the maintenance cycle of batch home appliances provided in the above-described embodiments of the present invention.

[0094] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage system located remotely from the aforementioned processor.

[0095] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware systems.

[0096] The method provided in this invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc. No limitation is made herein; any electronic device that can implement this invention falls within the protection scope of this invention.

[0097] For system / electronic device embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be found in the description of the method embodiments.

[0098] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0099] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0100] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0101] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for determining the maintenance cycle of batch home appliances, characterized in that, Includes the following steps: The batch common feature coefficients are determined based on the batch common features of multiple current home appliances of the current type; among which, the batch common features include: brand consistency feature, model consistency feature, installation time difference feature, usage scenario consistency feature, and average daily usage time difference feature; Determine the baseline maintenance cycle based on historical fault information of the current type of home appliance; The batch loss coordination coefficient is determined based on the batch common characteristic coefficient and the average daily usage time of the current home appliances; The loss correction maintenance cycle is determined based on the baseline maintenance cycle and the batch loss coordination coefficient. The service matching factor is determined based on the number of available service personnel on the platform, the number of home appliances maintained by a single service personnel in a single visit, the current total number of home appliances, and the wear and tear correction maintenance cycle. The service correction and maintenance cycle is determined based on the service adaptation factor and the loss correction and maintenance cycle. The probability of sudden failure is determined based on the baseline maintenance cycle, the service correction maintenance cycle, the total number of home appliances, and the preset basic fault risk threshold. Determine whether the probability of the sudden failure is greater than the preset basic failure risk threshold; If so, the target maintenance cycle is calculated iteratively based on the Poisson distribution and preset adjustment coefficients until the probability of multiple home appliances failing is less than the preset demand risk threshold.

2. The method as described in claim 1, characterized in that, The batch common feature coefficient The calculation formula is: in, This indicates brand consistency. Indicates model consistency characteristics. Indicates the installation time difference characteristic. This indicates consistency in usage scenarios. This indicates the characteristic of the difference in average daily usage time. These represent the corresponding preset weight coefficients.

3. The method as described in claim 2, characterized in that, The historical fault information of home appliances includes: the number of historical fault home appliance samples whose similarity to the batch common features is greater than or equal to a preset threshold, and the number of samples with historical faults. The time of failure of a historical home appliance sample and the first Average daily usage time of a sample of historical home appliances; The benchmark maintenance cycle The calculation formula is: in, M This represents the number of historical fault samples of home appliances whose similarity to the common features of the batch is greater than or equal to a preset threshold. Indicates the first The time of failure for a sample of historical home appliances; Indicates the first Average daily usage time of a sample of historical home appliances; This represents the fail-safe factor.

4. The method as described in claim 3, characterized in that, The batch loss coordination coefficient The calculation formula is: in, This indicates the average daily usage time of the current home appliance, in hours.

5. The method as described in claim 4, characterized in that, The correction and maintenance cycle The calculation formula is: .

6. The method as described in claim 1, characterized in that, The service adaptation factor The calculation formula is: in, This indicates the number of available service personnel on the platform. This indicates the number of home appliances maintained by a single service personnel in a single visit. This indicates the total number of home appliances currently in use.

7. The method as described in claim 1, characterized in that, The service repair and maintenance cycle The calculation formula is: .

8. A system for determining the maintenance cycle of bulk home appliances, characterized in that, include: The first determining module is used to determine the batch common feature coefficients based on the batch common features of multiple current home appliances of the current type; wherein, the batch common features include: brand consistency feature, model consistency feature, installation time difference feature, usage scenario consistency feature, and average daily usage time difference feature; The second determining module is used to determine the baseline maintenance cycle based on historical fault information of the current type of home appliance. The third determining module is used to determine the batch loss coordination coefficient based on the batch common characteristic coefficient and the average daily usage time of the current home appliances; The first correction module is used to determine the loss correction maintenance cycle based on the baseline maintenance cycle and the batch loss coordination coefficient. The fourth determination module is used to determine the service adaptation factor based on the number of available service personnel on the platform, the number of home appliances maintained by a single service personnel in a single visit, the current total number of home appliances, and the wear and tear correction maintenance cycle. The second correction module is used to determine the service correction and maintenance cycle based on the service adaptation factor and the loss correction and maintenance cycle. The calculation module is used to determine the probability of sudden failure based on the baseline maintenance cycle, the service correction maintenance cycle, the total number of home appliances, and the preset basic fault risk threshold. The judgment module is used to determine whether the probability of the sudden failure is greater than the preset basic failure risk threshold. The iteration module is used to iteratively calculate the target maintenance cycle based on the Poisson distribution and preset adjustment coefficients if the probability of multiple home appliances failing is less than the preset demand risk threshold.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for determining the maintenance cycle of batch home appliances as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for determining the maintenance cycle of batch home appliances as described in any one of claims 1 to 7.

Images