# A Workload-Balancing Order Dispatch Scheme for O2O Food Delivery with Order Splitting Choice

^{*}

## Abstract

**:**

## 1. Introduction

- For this delivery service case, we found out that the assignment rule of new orders proposed by Steever et al. [6] is less efficient in real-life, especially for its uneven workload distribution. Consequently, we offer a three-stage order dispatch scheme, namely, pseudo-assign first, re-route second, and courier selection last, showing an impressive performance on balancing courier’s workloads, without needing to compromise on delivery efficiency. In addition, this new scheme attempts to handle the dynamic and uncertain issues by considering the responsiveness to future demands and robustness to uncertainty.
- To illustrate the feasibility and applicability of this new service mode, we have conducted extensive experiments, which involve the influence of customers’ preference of split delivery on the delivery efficiency and decision of economical employments under distribution-known demands, casting several vital managerial policies. These conclusions would make food delivery platforms more adaptive and profitable, if they implemented this new O2O business model, as competition in O2O food delivery gets fiercer, and higher customer satisfaction is badly needed.

## 2. Literature Review

## 3. Problem Definition

#### 3.1. Problem Statement

- (1)
- The courier can not refuse the order assigned by the delivery platform. In other words, once a courier is assigned to the new order, this assignment does not change anymore.
- (2)
- The courier has no fixed initial position, as well as the end position. When the courier finishes the current orders, he will wait in suit for the new order.
- (3)
- Order delay is allowed. Specifically, it can be later than the promised delivery time.

#### 3.2. Desired Order Dispatch Scheme

**Delivery in time.**The first and foremost goal is timely delivery. Since the delivery time seriously affects the quality of customer service, and that meeting customers’ needs is the foundation of the existence of this O2O business model, we need to do our utmost to guarantee on-time delivery. Therefore, delay rate (DR), the ratio of delayed orders to the total number of orders, is used as the primary measure to evaluate the performance of a designed order dispatch scheme. To further measure the customer satisfaction, we can also apply the average late time (ALT) of delayed orders and average early time (AET) of orders delivered early. However, since the negative effect of late delivery is much stronger than the positive effect of an early delivery, the importance of AET is much lower than that of ALT [28].

**Workload balance.**Since a large variety in the workload will cause some ill consequences (e.g., low job satisfaction and high employee turnover), we are supposed to balance the couriers’ workload. Although there are many indicators to measure the workload balance, such as the standard deviation of the driving distance [25] and lexicographic minimax approach [29], the courier pays more attention to his salary, which is directly related to the number of orders completed. Therefore, we utilize the standard deviation of the numbers of assigned orders of the couriers to represent the level of unbalance of the couriers’ workload in this study, denoted as the unbalance index–standard deviation (UIsd) in the following discussions. The smaller it is, the more balanced workload is, and the designed order dispatch scheme is more practical in real-life applications.

**Responsiveness to future demand.**When making the order dispatch decisions, we should consider satisfying the current demand quickly enough, as well as its impacts on the response to future orders. In other words, we do not expect the order allocation to always be optimal at the current decision point, but has a good performance (such as lower DR and UIsd) in the long-term.

**Robustness to uncertainty.**Various uncertain factors exist in the whole operation process of the O2O food delivery service, such as the time required by the restaurants to prepare the food and for the couriers to travel on the road. Although we can take advantage of historical data or predicted information when making the order dispatch, there is constantly a gap between them and what happens in real-life. Therefore, we hope that our solutions are robust to these uncertainties. In other words, when the actual time required in these operations changes in a small range, the designed order dispatch scheme still performs well.

## 4. Proposed Order Dispatch Scheme

#### 4.1. Pseudo-Assign and Re-Route

#### 4.2. Measurement of Responsiveness to Future Demand

Algorithm 1: Measurement of responsiveness of couriers to the new order. |

Algorithm 2: The trade-off between current optimal solution and future demand. |

#### 4.3. The Whole Process of Courier Selection

Algorithm 3: The whole process of order dispatch scheme. |

## 5. Experimental Analyses and Managerial Insights

#### 5.1. Parameter and Dataset Settings

#### 5.2. Performance of the Proposed Scheme

#### 5.3. Sensitivity Analyses

#### 5.3.1. The Impact of the Split Order Ratio

#### 5.3.2. The Effect of the Number of Couriers

#### 5.3.3. Influence of Uncertainty on Delivery Efficiency

#### 5.4. Managerial Insights

## 6. Conclusions

## Supplementary Materials

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Roh, M.; Park, K. Adoption of O2O food delivery services in South Korea: The moderating role of moral obligation in meal preparation. Int. J. Inf. Manag.
**2019**, 47, 262–273. [Google Scholar] [CrossRef] - Chen, J.F.; Wang, L.; Wang, S.; Wang, X.; Ren, H. An effective matching algorithm with adaptive tie-breaking strategy for online food delivery problem. Complex Intell. Syst.
**2021**, 1–22. [Google Scholar] [CrossRef] - Liu, Y. An optimization-driven dynamic vehicle routing algorithm for on-demand meal delivery using drones. Comput. Oper. Res.
**2019**, 111, 1–20. [Google Scholar] [CrossRef] - Cosmi, M.; Oriolo, G.; Piccialli, V.; Ventura, P. Single courier single restaurant meal delivery (without routing). Oper. Res. Lett.
**2019**, 47, 537–541. [Google Scholar] [CrossRef] - Dsouza, D.; Sharma, D. Online food delivery portals during COVID-19 times: An analysis of changing consumer behavior and expectations. Int. J. Inov. Sci.
**2020**, 13, 218–232. [Google Scholar] [CrossRef] - Steever, Z.; Karwan, M.; Murray, C. Dynamic courier routing for a food delivery service. Comput. Oper. Res.
**2019**, 107, 173–188. [Google Scholar] [CrossRef] - Shen, C.W.; Chen, M.; Wang, C.C. Analyzing the trend of O2O commerce by bilingual text mining on social media. Comput. Hum. Behav.
**2019**, 101, 474–483. [Google Scholar] [CrossRef] - Nasrollahi, M.; Fathi, M.R.; Sanouni, H.R.; Sobhani, S.M.; Behrooz, A. Impact of coercive and non-coercive environmental supply chain sustainability drivers on supply chain performance: Mediation role of monitoring and collaboration. Int. J. Sustain. Eng.
**2021**, 14, 98–106. [Google Scholar] [CrossRef] - Nasrollahi, M.; Fathi, M.R.; Sobhani, S.M.; Khosravi, A.; Noorbakhsh, A. Modeling resilient supplier selection criteria in desalination supply chain based on fuzzy DEMATEL and ISM. Int. J. Manag. Sci. Eng. Manag.
**2021**, 16, 264–278. [Google Scholar] [CrossRef] - Safari, H.; Etezadi, S.; Moradi-Moghadam, M.; Fathi, M.R. Maturity evaluation of supply chain procedures by combining SCOR and PST models. Int. J. Process Manag. Benchmark.
**2021**, 11, 707–724. [Google Scholar] [CrossRef] - Fathi, M.R.; Nasrollahi, M.; Zamanian, A. Mathematical modeling of sustainable supply chain networks under uncertainty and solving it using metaheuristic algorithms. Ind. Manag. J.
**2020**, 11, 621–652. [Google Scholar] - Jiang, G.; Wang, Q.; Wang, K.; Zhang, Q.; Zhou, J. A novel closed-loop supply chain network design considering enterprise profit and service level. Sustainability
**2020**, 12, 544. [Google Scholar] [CrossRef][Green Version] - Ulmer, M.W.; Thomas, B.W.; Campbell, A.M.; Woyak, N. The restaurant meal delivery problem: Dynamic pickup and delivery with deadlines and random ready times. Transp. Sci.
**2021**, 55, 75–100. [Google Scholar] [CrossRef] - Yildiz, B.; Savelsbergh, M. Provably high-quality solutions for the meal delivery routing problem. Transp. Sci.
**2019**, 53, 1372–1388. [Google Scholar] [CrossRef][Green Version] - Seghezzi, A.; Winkenbach, M.; Mangiaracina, R. On-demand food delivery: A systematic literature review. Int. J. Logist. Manag.
**2021**, 32, 1334–1355. [Google Scholar] [CrossRef] - Allen, J.; Piecyk, M.; Cherrett, T.; Juhari, M.N.; McLeod, F.; Piotrowska, M.; Bates, O.; Bektas, T.; Cheliotis, K.; Friday, A.; et al. Understanding the transport and CO
_{2}impacts of on-demand meal deliveries: A London case study. Cities**2021**, 108, 102973. [Google Scholar] [CrossRef] - Psaraftis, H.N.; Wen, M.; Kontovas, C.A. Dynamic vehicle routing problems: Three decades and counting. Networks
**2016**, 67, 3–31. [Google Scholar] [CrossRef][Green Version] - Barkaoui, M. A co-evolutionary approach using information about future requests for dynamic vehicle routing problem with soft time windows. Memet. Comput.
**2018**, 10, 307–319. [Google Scholar] [CrossRef] - Hvattum, L.M.; Løkketangen, A.; Laporte, G. Solving a dynamic and stochastic vehicle routing problem with a sample scenario hedging heuristic. Transp. Sci.
**2006**, 40, 421–438. [Google Scholar] [CrossRef] - Ulmer, M.W.; Mattfeld, D.C.; Köster, F. Budgeting time for dynamic vehicle routing with stochastic customer requests. Transp. Sci.
**2018**, 52, 20–37. [Google Scholar] [CrossRef] - Ulmer, M.W.; Goodson, J.C.; Mattfeld, D.C.; Hennig, M. Offline-online approximate dynamic programming for dynamic vehicle routing with stochastic requests. Transp. Sci.
**2019**, 53, 185–202. [Google Scholar] [CrossRef][Green Version] - Abbatecola, L.; Fanti, M.P.; Pedroncelli, G.; Ukovich, W. A distributed cluster-based approach for pick-up services. IEEE Trans. Autom. Sci. Eng.
**2018**, 16, 960–971. [Google Scholar] [CrossRef] - Lesmana, N.S.; Zhang, X.; Bei, X. Balancing efficiency and fairness in on-demand ridesourcing. Adv. Neural Inf. Process. Syst.
**2019**, 32, 5309–5319. [Google Scholar] - Bozorgi-Amiri, A.; Khorsi, M. A dynamic multi-objective location-routing model for relief logistic planning under uncertainty on demand, travel time, and cost parameters. Int. J. Adv. Manuf. Technol.
**2016**, 85, 1633–1648. [Google Scholar] [CrossRef] - Benjamin, A.M.; Rahman, S.A.; Hashim, F. Driver’s workload analysis for waste collection vehicle routing problem. In Proceedings of the 2015 International Symposium on Mathematical Sciences and Computing Research (iSMSC), Ipoh, Malaysia, 19–20 May 2015; Publishing House: Ipoh, Malaysia, 2015; pp. 392–395. [Google Scholar]
- Matl, P.; Hartl, R.F.; Vidal, T. Workload equity in vehicle routing: The impact of alternative workload resources. Comput. Oper. Res.
**2019**, 110, 116–129. [Google Scholar] [CrossRef][Green Version] - Keskinturk, T.; Yildirim, M.B. A genetic algorithm metaheuristic for bakery distribution vehicle routing problem with load balancing. In Proceedings of the 2011 International Symposium on Innovations in Intelligent Systems and Applications, Istanbul, Turkey, 15–18 June 2011; IEEE: Istanbul, Turkey, 2011; pp. 287–291. [Google Scholar]
- Mao, W.; Ming, L.; Rong, Y.; Tang, C.S.; Zheng, H. Faster Deliveries and Smarter Order Assignments for an On-Demand Meal Delivery Platform. SSRN 3469015. 2019. Available online: https://anderson-review.ucla.edu/wp-content/uploads/2021/03/Mao-et-al_OnlineMealDelivery_2019_SSRN-id3469015.pdf (accessed on 25 November 2021).
- Lehuédé, F.; Péton, O.; Tricoire, F. A lexicographic minimax approach to the vehicle routing problem with route balancing. Eur. J. Oper. Res.
**2020**, 282, 129–147. [Google Scholar] [CrossRef] - Boscoe, F.P.; Henry, K.A.; Zdeb, M.S. A nationwide comparison of driving distance versus straight-line distance to hospitals. Prof. Geogr.
**2012**, 64, 188–196. [Google Scholar] [CrossRef][Green Version] - Zou, G.; Tang, J.; Yilmaz, L.; Kong, X. Online food ordering delivery strategies based on deep reinforcement learning. Appl. Intell.
**2021**, 1–13. [Google Scholar] [CrossRef]

**Figure 5.**Comparison of three algorithms in ALT under different split order ratio (Datasets 1 to 4).

**Figure 6.**Comparison of three algorithms in UIsd under different split order ratios (Datasets 1 to 4).

**Figure 9.**The impact of the difference between the expected and actual order preparation time on delivery efficiency.

Parameter | Explanation |
---|---|

V | Set of available couriers. |

R | Set of available restaurants. |

C | Set of customers whose orders have not been serviced. |

${C}_{v}$ | Set of unserved customers who are assigned to courier (v). |

${I}_{c}$ | Set of restaurants to which customer ($c\in C$) placed the order. |

${c}^{new},{I}_{{c}^{new}}$ | A new customer and the set of restaurants ordered by this customer. |

${\Delta}_{v}^{0}$ | Initial node (position) for courier ($v\in V$). |

$\gamma $ | A penalty coefficient $(\gamma \ge 0)$. |

${t}_{c}^{max}$ | Promised delivery time for customer (c). |

${t}_{i}^{min}$ | Estimated time for restaurant ($i\in R$) to finish the preparation of food. |

${\delta}^{max}$ | Maximum tolerable delay time. |

Notations | Explanation |
---|---|

$\widehat{{t}_{i}}$ | The future time point to be considered, $i\in [-2,-1,0,1,2]$. |

${t}_{now}$ | The current time when a new order places. |

${\mu}_{a}$ | The mean customer interarrival time. |

${\sigma}_{a}$ | The standard deviation associated with ${\mu}_{a}$. |

${t}_{r}$ | The average order preparation time of restaurant (r). |

${P}_{r}$ | The proportion of the customer placing an order in the restaurant (r). |

${\phi}_{i}^{v}$ | Set of intermediate nodes in the path of each courier (v), whose estimated arrival time is between $\widehat{{t}_{i}}$ and ${t}_{sc}^{max}$. |

$\eta $ | A positive integer. |

$nod{e}_{r,\widehat{{t}_{i}}}^{v}$ | The node where the courier (v) is closest to restaurant (r) at time $\widehat{{t}_{i}}$ |

$di{s}_{r,\widehat{{t}_{i}}}^{v}$ | The distance between $nod{e}_{r,\widehat{{t}_{i}}}^{v}$ and restaurant (r). |

$di{s}_{r,\widehat{{t}_{i}}}^{\eta v}$ | Mean distance of the $\eta $ nearest distances between the node in $\{nod{e}_{r,\widehat{{t}_{i}}}^{v},v\in V\}$ and restaurant (r). |

$StdDi{s}_{r,\widehat{{t}_{i}}}^{\eta v}$ | The standardized value of $di{s}_{r,\widehat{{t}_{i}}}^{\eta v}$ by $di{s}_{r,\widehat{{t}_{i}}}^{\eta v}/ma{x}_{r\in R}\left(di{s}_{r,\widehat{{t}_{i}}}^{\eta v}\right)$. |

${M}^{v}$ | The responsiveness of courier (v) to future orders. |

${\lambda}_{i}$ | Constant, ${\lambda}_{i}\in (0,1)$. |

${\mu}_{3}^{c}$ | The maximum emerging rate of customer orders. |

$\Delta t$ | The promised consumption time, from the time order was placed to the time the order arrived at customer destination. |

$\tau $ | The difference between the actual order preparation time and estimated order preparation time. |

Pre-Test Dataset 1 | Pre-Test Dataset 2 | Pre-Test Dataset 3 | |
---|---|---|---|

$\left|R\right|$ | 10 | 10 | 20 |

$\left|V\right|$ | 6 | 6 | 10 |

${\mu}_{3}^{c}$ | 10 | 10 | 24 |

Split prob. | 1 | 1 | 1 |

${t}_{v}^{{}^{\prime}}$ | 0 | 0 | 0 |

${t}_{v}^{{}^{\u2033}}$ | 4 | 4 | 4 |

$\Delta t$ (min) | 45 | 45 | 45 |

${t}_{r}$ (min) | $N(15,0)$ | $N(15,2.5)$ | $N(15,2.5)$ |

$\tau $ (min) | $N(0,5)$ | $N(0,1.5)$ | $N(0,1.5)$ |

Dataset 1 | Dataset 2 | Dataset 3 | Dataset 4 | Dataset 5 | Dataset 6 | |
---|---|---|---|---|---|---|

$\left|R\right|$ | 10 | 10 | 10 | 20 | 20 | 20 |

${\mu}_{3}^{c}$ | 10 | 10 | 10 | 24 | 24 | 24 |

${t}_{v}^{{}^{\prime}}$ | 0 | 0 | 0 | 0 | 0 | 0 |

${t}_{v}^{{}^{\u2033}}$ | 4 | 4 | 4 | 4 | 4 | 4 |

$\Delta t$ (min) | 45 | 45 | 45 | 45 | 45 | 45 |

${t}_{r}$ (min) | $N(15,2.5)$ | $N(15,2.5)$ | $N(15,2.5)$ | $N(15,2.5)$ | $N(15,2.5)$ | $N(15,2.5)$ |

$\tau $ (min) | $N(0,6)$ | $N(0,1.5)$ | $N(0,1.5)$ | $N(0,6)$ | $N(0,1.5)$ | $U(5,15)$ |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Wang, K.; Zhou, Y.; Zhang, L.
A Workload-Balancing Order Dispatch Scheme for O2O Food Delivery with Order Splitting Choice. *J. Theor. Appl. Electron. Commer. Res.* **2022**, *17*, 295-312.
https://doi.org/10.3390/jtaer17010015

**AMA Style**

Wang K, Zhou Y, Zhang L.
A Workload-Balancing Order Dispatch Scheme for O2O Food Delivery with Order Splitting Choice. *Journal of Theoretical and Applied Electronic Commerce Research*. 2022; 17(1):295-312.
https://doi.org/10.3390/jtaer17010015

**Chicago/Turabian Style**

Wang, Ke, Yulin Zhou, and Lingzhen Zhang.
2022. "A Workload-Balancing Order Dispatch Scheme for O2O Food Delivery with Order Splitting Choice" *Journal of Theoretical and Applied Electronic Commerce Research* 17, no. 1: 295-312.
https://doi.org/10.3390/jtaer17010015